ENGR296 - Final Project

</center>

Credit Card Fraud Detection

</center>

Name : Amit Padgaonkar

</center>

Student ID : 862188375

</center>

import numpy as np
import pandas as pd
import seaborn as sns
import warnings

warnings.filterwarnings('ignore')

df=pd.read_csv('creditcard.csv')
df.head()

	Time	V1	V2	V3	V4	V5	V6	V7	V8	V9	...	V21	V22	V23	V24	V25	V26	V27	V28	Amount	Class
0	0.0	-1.359807	-0.072781	2.536347	1.378155	-0.338321	0.462388	0.239599	0.098698	0.363787	...	-0.018307	0.277838	-0.110474	0.066928	0.128539	-0.189115	0.133558	-0.021053	149.62	0
1	0.0	1.191857	0.266151	0.166480	0.448154	0.060018	-0.082361	-0.078803	0.085102	-0.255425	...	-0.225775	-0.638672	0.101288	-0.339846	0.167170	0.125895	-0.008983	0.014724	2.69	0
2	1.0	-1.358354	-1.340163	1.773209	0.379780	-0.503198	1.800499	0.791461	0.247676	-1.514654	...	0.247998	0.771679	0.909412	-0.689281	-0.327642	-0.139097	-0.055353	-0.059752	378.66	0
3	1.0	-0.966272	-0.185226	1.792993	-0.863291	-0.010309	1.247203	0.237609	0.377436	-1.387024	...	-0.108300	0.005274	-0.190321	-1.175575	0.647376	-0.221929	0.062723	0.061458	123.50	0
4	2.0	-1.158233	0.877737	1.548718	0.403034	-0.407193	0.095921	0.592941	-0.270533	0.817739	...	-0.009431	0.798278	-0.137458	0.141267	-0.206010	0.502292	0.219422	0.215153	69.99	0

5 rows × 31 columns

Data Exploration

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 284807 entries, 0 to 284806
Data columns (total 31 columns):
 #   Column  Non-Null Count   Dtype  
---  ------  --------------   -----  
 0   Time    284807 non-null  float64
 1   V1      284807 non-null  float64
 2   V2      284807 non-null  float64
 3   V3      284807 non-null  float64
 4   V4      284807 non-null  float64
 5   V5      284807 non-null  float64
 6   V6      284807 non-null  float64
 7   V7      284807 non-null  float64
 8   V8      284807 non-null  float64
 9   V9      284807 non-null  float64
 10  V10     284807 non-null  float64
 11  V11     284807 non-null  float64
 12  V12     284807 non-null  float64
 13  V13     284807 non-null  float64
 14  V14     284807 non-null  float64
 15  V15     284807 non-null  float64
 16  V16     284807 non-null  float64
 17  V17     284807 non-null  float64
 18  V18     284807 non-null  float64
 19  V19     284807 non-null  float64
 20  V20     284807 non-null  float64
 21  V21     284807 non-null  float64
 22  V22     284807 non-null  float64
 23  V23     284807 non-null  float64
 24  V24     284807 non-null  float64
 25  V25     284807 non-null  float64
 26  V26     284807 non-null  float64
 27  V27     284807 non-null  float64
 28  V28     284807 non-null  float64
 29  Amount  284807 non-null  float64
 30  Class   284807 non-null  int64  
dtypes: float64(30), int64(1)
memory usage: 67.4 MB

All our columns are Numeric, which means we don't need any Categorical Label -> Numeric data conversion

#Calculating the % of Null values in each column
df.isnull().sum()/df.shape[0]

Time      0.0
V1        0.0
V2        0.0
V3        0.0
V4        0.0
V5        0.0
V6        0.0
V7        0.0
V8        0.0
V9        0.0
V10       0.0
V11       0.0
V12       0.0
V13       0.0
V14       0.0
V15       0.0
V16       0.0
V17       0.0
V18       0.0
V19       0.0
V20       0.0
V21       0.0
V22       0.0
V23       0.0
V24       0.0
V25       0.0
V26       0.0
V27       0.0
V28       0.0
Amount    0.0
Class     0.0
dtype: float64

There are no NULL values

df.shape

(284807, 31)

Checking obvious correlations in the data

df.corr()
df.corr().style.set_precision(3)

	Time	V1	V2	V3	V4	V5	V6	V7	V8	V9	V10	V11	V12	V13	V14	V15	V16	V17	V18	V19	V20	V21	V22	V23	V24	V25	V26	V27	V28	Amount	Class
Time	1.000	0.117	-0.011	-0.420	-0.105	0.173	-0.063	0.085	-0.037	-0.009	0.031	-0.248	0.124	-0.066	-0.099	-0.183	0.012	-0.073	0.090	0.029	-0.051	0.045	0.144	0.051	-0.016	-0.233	-0.041	-0.005	-0.009	-0.011	-0.012
V1	0.117	1.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	-0.228	-0.101
V2	-0.011	0.000	1.000	0.000	-0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	-0.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	-0.531	0.091
V3	-0.420	-0.000	0.000	1.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.211	-0.193
V4	-0.105	0.000	-0.000	-0.000	1.000	-0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.099	0.133
V5	0.173	0.000	-0.000	-0.000	-0.000	1.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.386	-0.095
V6	-0.063	0.000	0.000	0.000	-0.000	0.000	1.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	0.216	-0.044
V7	0.085	0.000	0.000	0.000	0.000	-0.000	0.000	1.000	-0.000	0.000	0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	-0.000	0.397	-0.187
V8	-0.037	-0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	1.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	-0.000	-0.103	0.020
V9	-0.009	0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	0.000	1.000	-0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	-0.000	0.000	-0.044	-0.098
V10	0.031	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	1.000	0.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.102	-0.217
V11	-0.248	0.000	0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	1.000	0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.155
V12	0.124	0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	1.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	-0.010	-0.261
V13	-0.066	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	1.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	0.005	-0.005
V14	-0.099	0.000	-0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	0.000	1.000	0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	0.034	-0.303
V15	-0.183	-0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	1.000	0.000	0.000	0.000	-0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.003	-0.004
V16	0.012	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	1.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.004	-0.197
V17	-0.073	-0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	1.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	0.007	-0.326
V18	0.090	0.000	0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	1.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.000	0.000	0.036	-0.111
V19	0.029	0.000	0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	-0.000	-0.000	1.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	-0.056	0.035
V20	-0.051	0.000	-0.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	-0.000	-0.000	0.000	1.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.339	0.020
V21	0.045	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	-0.000	1.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.106	0.040
V22	0.144	0.000	0.000	0.000	0.000	0.000	-0.000	-0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	1.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.065	0.001
V23	0.051	0.000	0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	0.000	0.000	0.000	0.000	1.000	-0.000	-0.000	0.000	0.000	0.000	-0.113	-0.003
V24	-0.016	0.000	-0.000	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	0.000	0.000	0.000	-0.000	1.000	0.000	0.000	-0.000	-0.000	0.005	-0.007
V25	-0.233	-0.000	-0.000	0.000	0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	1.000	0.000	-0.000	0.000	-0.048	0.003
V26	-0.041	-0.000	0.000	0.000	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	0.000	0.000	0.000	1.000	-0.000	-0.000	-0.003	0.004
V27	-0.005	0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	0.000	0.000	0.000	-0.000	-0.000	-0.000	-0.000	0.000	-0.000	-0.000	-0.000	1.000	-0.000	0.029	0.018
V28	-0.009	0.000	-0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	-0.000	-0.000	0.000	0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	0.000	-0.000	0.000	-0.000	0.000	-0.000	-0.000	1.000	0.010	0.010
Amount	-0.011	-0.228	-0.531	-0.211	0.099	-0.386	0.216	0.397	-0.103	-0.044	-0.102	0.000	-0.010	0.005	0.034	-0.003	-0.004	0.007	0.036	-0.056	0.339	0.106	-0.065	-0.113	0.005	-0.048	-0.003	0.029	0.010	1.000	0.006
Class	-0.012	-0.101	0.091	-0.193	0.133	-0.095	-0.044	-0.187	0.020	-0.098	-0.217	0.155	-0.261	-0.005	-0.303	-0.004	-0.197	-0.326	-0.111	0.035	0.020	0.040	0.001	-0.003	-0.007	0.003	0.004	0.018	0.010	0.006	1.000

Plotting the heatmap of data to see if there any obvious correlations

import matplotlib.pyplot as plt

figure = plt.figure(figsize=(20,20))
figure.set_facecolor('w')
sns.heatmap(df.corr(),annot=False,linewidths=0.01,cmap='coolwarm')
plt.show()

We can see from the correlation data and heatmap, we can see that the V13,V15,V22,V23,V24,V25,V26 have very little correlation with our target variable Class

import seaborn as sns
sns.set(rc={'figure.facecolor':'white'})
sns.set_style("whitegrid", {'axes.grid' : False})
sns.pairplot(df,vars = ['Class', 'Amount', 'Time','V1','V2'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dc6bf3da48>

sns.pairplot(df,vars = ['Class','V3', 'V4', 'V5','V6'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dc6c7d2688>

sns.pairplot(df,vars = ['Class','V7', 'V8', 'V9','V10'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dc6b5ecf08>

sns.pairplot(df,vars = ['Class','V11', 'V12', 'V13','V14'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dc7c607948>

sns.pairplot(df,vars = ['Class','V15', 'V16', 'V17','V18'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dbf2d05308>

sns.pairplot(df,vars = ['Class','V19', 'V20', 'V21','V22'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dbf2cecd88>

sns.pairplot(df,vars = ['Class','V23', 'V24', 'V25','V26'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dc6ac88348>

sns.pairplot(df,vars = ['Class','V27', 'V28'], hue = 'Class')

<seaborn.axisgrid.PairGrid at 0x1dc6c217f08>

import matplotlib.pyplot as plt

figure = plt.figure(figsize=(15,6))
figure.set_facecolor('w')
ax = df.corr()['Class'].sort_values().plot(kind='bar',color='b',grid=False)
ax.set_facecolor('w')

We will drop the columns V13,V15,V22,V23,V25,V26 which have no correlation and may not add value to our classification model

print("The range of Amount column min = "+ str(df['Amount'].min()) +" max = "+ str(df['Amount'].max()))

The range of Amount column min = 0.0 max = 25691.16

The range of Amount is high, we should convert it to log scale.

Checking the distribution of values in the target column Class

df["Class"].value_counts()

0    284315
1       492
Name: Class, dtype: int64

df["Class"].value_counts()/df.shape[0]

0    0.998273
1    0.001727
Name: Class, dtype: float64

fig, ax = plt.subplots(figsize=(12,30))
sns.countplot(ax=ax, x='Class', data=df)

<matplotlib.axes._subplots.AxesSubplot at 0x1db86b5ce08>

Values for the minority class which indicate fraudulent transacations are only 0.1727 % of the total data. The fraud class bar on the barchart is barely visible. This is clearly a SEVERLY IMBALANCED dataset.

Data Visualisation in 2-dimensional space

#For sampling, we will get all the samples from fraud class and only a fraction of legitimate transactions
fraud_instances = len(df[df["Class"] == 1])
fraud_data = df[df["Class"] == 1]
legitimate_data = df[df["Class"] == 0]
subset_legitimate_data = legitimate_data.sample(fraud_instances*12) #Only about 2% of legitimate transactions
fraud_data.reset_index(drop=True, inplace=True)
subset_legitimate_data.reset_index(drop=True, inplace=True)
sample_data = pd.concat([fraud_data,subset_legitimate_data])

X_visualization = sample_data.drop('Class', axis=1)
y_visualization = sample_data['Class']

#Applying the standard scaler that we are going to use later
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()

X_visualization=scaler.fit_transform(X_visualization)

#Run TSNE dimensionality reduction
from sklearn.manifold import TSNE
X_reduced_dim = TSNE(n_components=2, random_state=42).fit_transform(X_visualization)

# t-SNE scatter plot
import matplotlib.patches as mpatches

f, ax = plt.subplots(figsize=(24,16))


legitimate = mpatches.Patch(color='blue', label='Legitimate')
fraudulent = mpatches.Patch(color='red', label='Fraudulent')

ax.scatter(X_reduced_dim[:,0], X_reduced_dim[:,1], c=(y_visualization == 0), cmap='coolwarm', label='No Fraud', linewidths=2)
ax.scatter(X_reduced_dim[:,0], X_reduced_dim[:,1], c=(y_visualization == 1), cmap='coolwarm', label='Fraud', linewidths=2)
ax.set_title('Visualizing the transformed CC Fraud Dataset', fontsize=14)

ax.grid(False)
ax.set_facecolor('w')

ax.legend(handles=[legitimate, fraudulent])

<matplotlib.legend.Legend at 0x1dbda24b388>

Data Preprocessing

Let's drop the columns that have very little correlation

df.drop(columns=['V22','V25','V26'],inplace=True, axis=1)

Convert the amount to log scale

df['Amount'] = np.log(df['Amount'] + 0.001)

The number of minority class (positive) samples are very rare (only 492). While splitting the dataset, let us do it in a way that maintains the same class distribution in each subset. We will do it through STRATIFIED SAMPLING by using target variable to control the sampling process.

#Class is the target variable
X=df.loc[:, df.columns != 'Class']
y=df[['Class']]

X.shape, y.shape

((284807, 27), (284807, 1))

# split into train/test sets with same class ratio
from sklearn.model_selection import train_test_split
trainX, testX, trainy, testy = train_test_split(X, y, test_size=0.3, random_state=100,stratify=y)

train_0, train_1 = len(trainy[trainy['Class']==0]), len(trainy[trainy['Class']==1])
test_0, test_1 = len(testy[testy['Class']==0]), len(testy[testy['Class']==1])
print('Checking the Split -->Train: 0=%d, 1=%d, Test: 0=%d, 1=%d' % (train_0, train_1, test_0, test_1))

Checking the Split -->Train: 0=199020, 1=344, Test: 0=85295, 1=148

Split looks good

One approach to addressing imbalanced datasets is to oversample the minority class. New examples can be synthesized from the existing examples. This is a type of data augmentation for the minority class and is referred to as the Synthetic Minority Oversampling Technique (SMOTE). We will use a variant of SMOTE based on Support Vector Machine(SVM) called SVMSMOTE

from collections import Counter

counter = Counter(trainy['Class'])
print(counter)

Counter({0: 199020, 1: 344})

from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()

trainX=scaler.fit_transform(trainX)
testX=scaler.transform(testX)

from imblearn.over_sampling import SVMSMOTE
oversample = SVMSMOTE(sampling_strategy=1)
trainX_scaled, trainy_scaled = oversample.fit_resample(trainX, trainy)
counter = Counter(trainy_scaled['Class'])

print(counter)

Counter({0: 199020, 1: 199020})

fig, ax = plt.subplots(figsize=(12,12))
sns.countplot(ax=ax, x='Class',data=trainy_scaled)

<matplotlib.axes._subplots.AxesSubplot at 0x1db86ab3108>

Oversized sampling has brought the size of minority class sample to one third of that of majority class.

# Arrays for barplots

arr_table_labels = []
arr_barchart_labels = []
arr_accuracy = []
arr_precision = []
arr_recall = []
arr_f1score = []
arr_f2measure = []
arr_roc_auc = []
arr_tn = []
arr_fp = []
arr_fn = []
arr_tp = []

from sklearn.metrics import confusion_matrix
def plot_cm(ax,labels, predictions,color_index,title, p=0.5):
  cm_colors = ['BuGn','coolwarm','YlOrRd','YlOrBr','OrRd', 'PuRd', 'RdPu', 'BuPu']
  cm = confusion_matrix(labels, predictions > p)
  #plt.figure(figsize=(5,5))
  sns.heatmap(cm, annot=True, fmt="d",xticklabels = ["No", "Yes"] , yticklabels = ["legitimate", "Fraudulent"],cbar=False,cmap=cm_colors[color_index])
  ax.set_title(title.format(p))
  ax.set_ylabel('Actual label')
  ax.set_xlabel('Predicted label')
  ax.set_facecolor('w')

  print('Legitimate Transactions Detected (True Negatives): ', cm[0][0])
  arr_tn.append(cm[0][0])
  print('Legitimate Transactions Incorrectly Detected (False Positives): ', cm[0][1])
  arr_fp.append(cm[0][1])
  print('Fraudulent Transactions Missed (False Negatives): ', cm[1][0])
  arr_fn.append(cm[1][0])
  print('Fraudulent Transactions Detected (True Positives): ', cm[1][1])
  arr_tp.append(cm[1][1])
  print('Total Fraudulent Transactions in validation dataset: ', np.sum(cm[1]))

from sklearn import metrics
from sklearn.metrics import accuracy_score
from sklearn.metrics import precision_score
from sklearn.metrics import recall_score
from sklearn.metrics import f1_score
from sklearn.metrics import roc_auc_score
from sklearn.metrics import roc_curve
from sklearn.metrics import fbeta_score
from sklearn.metrics import auc
from sklearn.metrics import precision_recall_curve

def calculate_and_plot(model,color_index,title,label,algorithm):
    # predict crisp classes for test set
    yhat_classes = model.predict(testX)
    # predict probabilities for test set
    yhat_probs = model.predict_proba(testX)
    
    fig, (ax1, ax2) = plt.subplots(1, 2)
    fig.set_size_inches(15, 6)
    
    calculate_and_print_scores(testy, yhat_classes,yhat_probs[:, 1])

    plot_roc_auc(ax1,testy, yhat_probs[:, 1], label)
    
    plot_cm(ax2,testy, yhat_classes,color_index,title)
       
    create_barchart_label(label,algorithm)
    
    create_table_label(label,algorithm)

def calculate_and_print_scores(testy, yhat_classes, yhat_probs):
    # accuracy: (tp + tn) / (p + n)
    accuracy = accuracy_score(testy, yhat_classes)
    print('Accuracy: %f' % accuracy)
    arr_accuracy.append(accuracy)
    
    # precision tp / (tp + fp)
    precision = precision_score(testy, yhat_classes)
    print('Precision: %f' % precision)
    arr_precision.append(precision)
      
    # recall: tp / (tp + fn)
    recall = recall_score(testy, yhat_classes)
    print('Sensitivity AKA Recall: %f' % recall)
    arr_recall.append(recall)
    
    # f1: 2 tp / (2 tp + fp + fn)
    f1 = f1_score(testy, yhat_classes)
    print('F1 score: %f' % f1)
    arr_f1score.append(f1)
    
    #F2 Measure
    f = fbeta_score(testy, yhat_classes, beta=2.0)
    print('F2-Measure: %f' % f)
    arr_f2measure.append(f)
   
    # ROC AUC
    rocauc = roc_auc_score(testy, yhat_probs)
    print('ROC AUC: %f' % rocauc)
    arr_roc_auc.append(rocauc)

import matplotlib.pyplot as plt

def plot_roc_auc(ax,testy, yhat_probs,label):
    #Show ROC-AUC Plot
    #fig = plt.figure(figsize=(5,5))
    #fig.set_facecolor('w')
    # plot no skill roc curve
    ax.plot([0, 1], [0, 1], linestyle='--', color='red', label='No Skill')
    # calculate roc curve for model
    fpr, tpr, _ = roc_curve(testy, yhat_probs)
    # plot model roc curve
    ax.plot(fpr, tpr, marker='.',color='green', linewidth=4, label = label)
    # axis labels
    ax.set_xlabel('False Positive Rate')
    ax.set_xlim(0,1)
    ax.set_ylabel('True Positive Rate')
    ax.set_ylim(0,1)
    ax.set_xticks(np.arange(0, 1, step=0.1))
    ax.set_yticks(np.arange(0, 1, step=0.1))
    ax.set_facecolor('w')
    

    #fig.suptitle('Receiver Operating Characteristics (ROC) Curve', fontsize=20)
    # show the legend
    ax.legend()
    # show the plot

def create_barchart_label(label,algorithm):
    barchart_label = algorithm.upper() + " -> " + label
    arr_barchart_labels.append(barchart_label)

def create_table_label(label,algorithm):
    table_label = algorithm.upper() + " -> " + label
    arr_table_labels.append(table_label)

Creating ML Models

</center>

1. Ensemble Algorithms

Balanced Bagging Classifier Algorithm

from imblearn.ensemble import BalancedBaggingClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import BaggingClassifier
import warnings

warnings.filterwarnings("ignore", category=FutureWarning)

def create_bal_bg_classifier_model(n_estimators=1,max_samples=0.1,max_depth=100,sampling_strategy=0.1):
    clf = BalancedBaggingClassifier(DecisionTreeClassifier(max_depth=max_depth),max_samples=max_samples,n_estimators=n_estimators,sampling_strategy=sampling_strategy,replacement=True)
    return clf

clf_bbg_001 = create_bal_bg_classifier_model(100,max_samples=0.1,max_depth=100,sampling_strategy=0.01)
clf_bbg_001 = clf_bbg_001.fit(trainX, trainy)

calculate_and_plot(clf_bbg_001,0,'Sampling Straegy : 0.01','Sampling Straegy : 0.01','Bagging')

Accuracy: 0.999122
Precision: 0.710983
Sensitivity AKA Recall: 0.831081
F1 score: 0.766355
F2-Measure: 0.803922
ROC AUC: 0.937494
Legitimate Transactions Detected (True Negatives):  85245
Legitimate Transactions Incorrectly Detected (False Positives):  50
Fraudulent Transactions Missed (False Negatives):  25
Fraudulent Transactions Detected (True Positives):  123
Total Fraudulent Transactions in validation dataset:  148

clf_bbg_01 = create_bal_bg_classifier_model(100,max_samples=0.1,max_depth=100,sampling_strategy=0.1)
clf_bbg_01 = clf_bbg_01.fit(trainX, trainy)

calculate_and_plot(clf_bbg_01,1,'Sampling Straegy : 0.1','Sampling Straegy : 0.1','Bagging')

Accuracy: 0.996781
Precision: 0.330667
Sensitivity AKA Recall: 0.837838
F1 score: 0.474187
F2-Measure: 0.641158
ROC AUC: 0.945874
Legitimate Transactions Detected (True Negatives):  85044
Legitimate Transactions Incorrectly Detected (False Positives):  251
Fraudulent Transactions Missed (False Negatives):  24
Fraudulent Transactions Detected (True Positives):  124
Total Fraudulent Transactions in validation dataset:  148

clf_bbg_1 = create_bal_bg_classifier_model(100,max_samples=0.1,max_depth=100,sampling_strategy=1)
clf_bbg_1.fit(trainX, trainy)

BalancedBaggingClassifier(base_estimator=DecisionTreeClassifier(max_depth=100),
                          max_samples=0.1, n_estimators=100, replacement=True,
                          sampling_strategy=1)

calculate_and_plot(clf_bbg_1,2,'Sampling Straegy : 1','Sampling Straegy : 1','Bagging')

Accuracy: 0.975235
Precision: 0.057951
Sensitivity AKA Recall: 0.871622
F1 score: 0.108677
F2-Measure: 0.228886
ROC AUC: 0.961538
Legitimate Transactions Detected (True Negatives):  83198
Legitimate Transactions Incorrectly Detected (False Positives):  2097
Fraudulent Transactions Missed (False Negatives):  19
Fraudulent Transactions Detected (True Positives):  129
Total Fraudulent Transactions in validation dataset:  148

Random Forest Classifier with Class Weighting

from sklearn.ensemble import RandomForestClassifier

def create_rf_classifier_model(n_estimators=100,max_samples=0.1,max_depth=10,class_weight='balanced_subsample'):
    clf = RandomForestClassifier(max_depth=max_depth,max_samples=max_samples,n_estimators=n_estimators,class_weight=class_weight)
    return clf

clf_rf_1 =  create_rf_classifier_model(100,max_samples=0.2,max_depth=15,class_weight = {0: 1, 1: 1})
clf_rf_1 = clf_rf_1.fit(trainX, trainy)

calculate_and_plot(clf_rf_1,0,'Class Weight 1 (Equal)','Class Weight 1','Random Forest')

Accuracy: 0.999356
Precision: 0.884298
Sensitivity AKA Recall: 0.722973
F1 score: 0.795539
F2-Measure: 0.750351
ROC AUC: 0.944044
Legitimate Transactions Detected (True Negatives):  85281
Legitimate Transactions Incorrectly Detected (False Positives):  14
Fraudulent Transactions Missed (False Negatives):  41
Fraudulent Transactions Detected (True Positives):  107
Total Fraudulent Transactions in validation dataset:  148

clf_rf_100 =  create_rf_classifier_model(100,max_samples=0.2,max_depth=15,class_weight = {0:1, 1:100})
clf_rf_100 = clf_rf_100.fit(trainX, trainy)

calculate_and_plot(clf_rf_100,1,'Class Weight 100','Class Weight 100','Random Forest')

Accuracy: 0.995681
Precision: 0.266385
Sensitivity AKA Recall: 0.851351
F1 score: 0.405797
F2-Measure: 0.591549
ROC AUC: 0.957426
Legitimate Transactions Detected (True Negatives):  84948
Legitimate Transactions Incorrectly Detected (False Positives):  347
Fraudulent Transactions Missed (False Negatives):  22
Fraudulent Transactions Detected (True Positives):  126
Total Fraudulent Transactions in validation dataset:  148

clf_rf_1000 =  create_rf_classifier_model(100,max_samples=0.2,max_depth=15,class_weight = {0:1, 1:1000})
clf_rf_1000 = clf_rf_100.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_rf_100,2,'Class Weight 1000','Class Weight 1000','Random Forest')

Accuracy: 0.980431
Precision: 0.072391
Sensitivity AKA Recall: 0.871622
F1 score: 0.133679
F2-Measure: 0.271693
ROC AUC: 0.972310
Legitimate Transactions Detected (True Negatives):  83642
Legitimate Transactions Incorrectly Detected (False Positives):  1653
Fraudulent Transactions Missed (False Negatives):  19
Fraudulent Transactions Detected (True Positives):  129
Total Fraudulent Transactions in validation dataset:  148

2. Cost Sensisitive Algorithms

Cost Sensitive Logistic Regression

from sklearn.linear_model import LogisticRegression

def create_cslr_classifier_model(class_weight):
    clf = LogisticRegression(solver='lbfgs',class_weight=class_weight)
    return clf

clf_lr_1 =  create_cslr_classifier_model([{0: 1, 1: 1}])
clf_lr_1 = clf_lr_1.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_lr_1,0,'Confusion Matrix with Weight 1 (equal weight) for Fraud Class','Class Weight 1','Logistic Regression')

Accuracy: 0.994640
Precision: 0.223214
Sensitivity AKA Recall: 0.844595
F1 score: 0.353107
F2-Measure: 0.542535
ROC AUC: 0.953544
Legitimate Transactions Detected (True Negatives):  84860
Legitimate Transactions Incorrectly Detected (False Positives):  435
Fraudulent Transactions Missed (False Negatives):  23
Fraudulent Transactions Detected (True Positives):  125
Total Fraudulent Transactions in validation dataset:  148

clf_lr_100 =  create_cslr_classifier_model([{0: 1, 1: 100}])
clf_lr_100 = clf_lr_100.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_lr_100,1,'Confusion Matrix with Weight 100 for Fraud Class','Class Weight 100','Logistic Regression')

Accuracy: 0.994640
Precision: 0.223214
Sensitivity AKA Recall: 0.844595
F1 score: 0.353107
F2-Measure: 0.542535
ROC AUC: 0.953544
Legitimate Transactions Detected (True Negatives):  84860
Legitimate Transactions Incorrectly Detected (False Positives):  435
Fraudulent Transactions Missed (False Negatives):  23
Fraudulent Transactions Detected (True Positives):  125
Total Fraudulent Transactions in validation dataset:  148

clf_lr_1000 =  create_cslr_classifier_model([{0: 1, 1: 1000}])
clf_lr_1000 = clf_lr_1000.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_lr_1000,2,'Confusion Matrix with Weight 1000 for Fraud Class','Class Weight 1000','Logistic Regression')

Accuracy: 0.994640
Precision: 0.223214
Sensitivity AKA Recall: 0.844595
F1 score: 0.353107
F2-Measure: 0.542535
ROC AUC: 0.953544
Legitimate Transactions Detected (True Negatives):  84860
Legitimate Transactions Incorrectly Detected (False Positives):  435
Fraudulent Transactions Missed (False Negatives):  23
Fraudulent Transactions Detected (True Positives):  125
Total Fraudulent Transactions in validation dataset:  148

Cost Sensitive Decision Trees

def create_csdt_classifier_model(class_weight,max_depth=None):
    model = DecisionTreeClassifier(max_depth=max_depth,class_weight=class_weight)
    return model

clf_dt_1 =  create_csdt_classifier_model({0: 1.0, 1: 1.0})
clf_dt_1 = clf_dt_1.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_dt_1,3,'Confusion Matrix with Weight 1 (equal weight) for Fraud Class','Class Weight 1','Decision Tree')

Accuracy: 0.999122
Precision: 0.708571
Sensitivity AKA Recall: 0.837838
F1 score: 0.767802
F2-Measure: 0.808344
ROC AUC: 0.918620
Legitimate Transactions Detected (True Negatives):  85244
Legitimate Transactions Incorrectly Detected (False Positives):  51
Fraudulent Transactions Missed (False Negatives):  24
Fraudulent Transactions Detected (True Positives):  124
Total Fraudulent Transactions in validation dataset:  148

clf_dt_10 =  create_csdt_classifier_model({0: 1.0, 1: 100.0})
clf_dt_10 = clf_dt_10.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_dt_10,4,'Confusion Matrix with Weight 100 for Fraud Class','Class Weight 100','Decision Tree')

Accuracy: 0.998912
Precision: 0.648649
Sensitivity AKA Recall: 0.810811
F1 score: 0.720721
F2-Measure: 0.772201
ROC AUC: 0.905024
Legitimate Transactions Detected (True Negatives):  85230
Legitimate Transactions Incorrectly Detected (False Positives):  65
Fraudulent Transactions Missed (False Negatives):  28
Fraudulent Transactions Detected (True Positives):  120
Total Fraudulent Transactions in validation dataset:  148

clf_dt_100 =  create_csdt_classifier_model({0: 1.0, 1: 1000.0})
clf_dt_100 = clf_dt_100.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_dt_100,5,'Confusion Matrix with Weight 1000 for Fraud Class','Class Weight 1000','Decision Tree')

Accuracy: 0.998876
Precision: 0.638298
Sensitivity AKA Recall: 0.810811
F1 score: 0.714286
F2-Measure: 0.769231
ROC AUC: 0.905007
Legitimate Transactions Detected (True Negatives):  85227
Legitimate Transactions Incorrectly Detected (False Positives):  68
Fraudulent Transactions Missed (False Negatives):  28
Fraudulent Transactions Detected (True Positives):  120
Total Fraudulent Transactions in validation dataset:  148

Cost Sensitive Gradient Boosting XGBoost

from xgboost import XGBClassifier

def create_xgb_classifier_model(scale_pos_weight):
    model = XGBClassifier(objective='binary:logistic',max_delta_step=5,scale_pos_weight=scale_pos_weight,max_depth=100)
    return model

clf_xg_1 =  create_xgb_classifier_model(1)
clf_xg_1 = clf_xg_1.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_xg_1,0,'Confusion Matrix with Weight 1 (equal weight) for Fraud Class','Class Weight 1','XGBoost')

Accuracy: 0.999345
Precision: 0.794872
Sensitivity AKA Recall: 0.837838
F1 score: 0.815789
F2-Measure: 0.828877
ROC AUC: 0.967556
Legitimate Transactions Detected (True Negatives):  85263
Legitimate Transactions Incorrectly Detected (False Positives):  32
Fraudulent Transactions Missed (False Negatives):  24
Fraudulent Transactions Detected (True Positives):  124
Total Fraudulent Transactions in validation dataset:  148

clf_xg_50 =  create_xgb_classifier_model(50)
clf_xg_50 = clf_xg_50.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_xg_50,1,'Confusion Matrix with Weight 50 for Fraud Class','Class Weight 50','XGBoost')

Accuracy: 0.999146
Precision: 0.716763
Sensitivity AKA Recall: 0.837838
F1 score: 0.772586
F2-Measure: 0.810458
ROC AUC: 0.974359
Legitimate Transactions Detected (True Negatives):  85246
Legitimate Transactions Incorrectly Detected (False Positives):  49
Fraudulent Transactions Missed (False Negatives):  24
Fraudulent Transactions Detected (True Positives):  124
Total Fraudulent Transactions in validation dataset:  148

clf_xg_100 =  create_xgb_classifier_model(500)
clf_xg_100 = clf_xg_100.fit(trainX_scaled, trainy_scaled)

calculate_and_plot(clf_xg_100,2,'Confusion Matrix with Weight 1000 for Fraud Class','Class Weight 1000','XGBoost')

Accuracy: 0.998853
Precision: 0.626263
Sensitivity AKA Recall: 0.837838
F1 score: 0.716763
F2-Measure: 0.784810
ROC AUC: 0.957629
Legitimate Transactions Detected (True Negatives):  85221
Legitimate Transactions Incorrectly Detected (False Positives):  74
Fraudulent Transactions Missed (False Negatives):  24
Fraudulent Transactions Detected (True Positives):  124
Total Fraudulent Transactions in validation dataset:  148

Cost Sensitive Artificial Neural Network

import tensorflow as tf
from tensorflow.keras import layers
from tensorflow import keras
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense,Dropout, Activation
from tensorflow.keras.optimizers import Adam,SGD
from keras.callbacks import EarlyStopping
from keras.callbacks import ModelCheckpoint

EPOCHS=200
BATCH_SIZE=1024

METRICS = [
      keras.metrics.TruePositives(name='tp'),
      keras.metrics.FalsePositives(name='fp'),
      keras.metrics.TrueNegatives(name='tn'),
      keras.metrics.FalseNegatives(name='fn'), 
      keras.metrics.BinaryAccuracy(name='accuracy'),
      keras.metrics.Precision(name='precision'),
      keras.metrics.Recall(name='recall'),
      keras.metrics.AUC(name='auc'),
      keras.metrics.SensitivityAtSpecificity(specificity=0.999,name='sensitivity'),
      keras.metrics.SpecificityAtSensitivity(sensitivity=0.8,name='specificity') 
]

def make_model(metrics = METRICS, act_function='relu',output_bias=None):
    model = Sequential()
    # define first visible layer 
    model.add(Dense(trainX_scaled.shape[1],input_dim=trainX_scaled.shape[1],activation=act_function))
    model.add(Dropout(0.4))
    #tanh has steeper gradients, so backprop is more effective in cost sensitive weights which affect backpropagation
    # define first hidden layer 
    model.add(Dense(32, activation=act_function)) #number of neurons are kept to the power of 2
    model.add(Dropout(0.4))
    # define second hidden layer
    model.add(Dense(32, activation=act_function)) #number of neurons are kept to the power of 2
    model.add(Dropout(0.4))
    # define output layer
    model.add(Dense(1, activation='sigmoid'))
    # Cost sensitive weights to punish the false negatives
    # define loss and optimizer
    model.compile(loss='binary_crossentropy', optimizer=keras.optimizers.Adam(lr=1e-3),metrics=METRICS)

    return model

#We can see that val_tp does not vary through the iterations a lot. However, by focussing on FPs, we are increasing the precision and user experience
es = EarlyStopping(monitor='val_precision',mode='max', verbose=1, restore_best_weights= True, patience=100)

#tanh has steeper gradients, so backprop is more effective in cost sensitive weights which affect backpropagation
model_no_weight = make_model(act_function='tanh')

#tanh has steeper gradients, so backprop is more effective in cost sensitive weights which affect backpropagation
model_5_weight = make_model(act_function='tanh')

#tanh has steeper gradients, so backprop is more effective in cost sensitive weights which affect backpropagation
model_500_weight = make_model(act_function='tanh')

Lets Run the 3 models

run_no_weight_history = model_no_weight.fit(x=trainX_scaled,y=trainy_scaled,class_weight={0:1,1:1},validation_data=(testX,testy),batch_size=BATCH_SIZE,epochs=EPOCHS,callbacks=[es])

Train on 398040 samples, validate on 85443 samples
Epoch 1/200
398040/398040 [==============================] - 6s 14us/sample - loss: 0.1244 - tp: 194086.0000 - fp: 16250.0000 - tn: 182770.0000 - fn: 4934.0000 - accuracy: 0.9468 - precision: 0.9227 - recall: 0.9752 - auc: 0.9927 - sensitivity: 0.6358 - specificity: 0.9968 - val_loss: 0.0204 - val_tp: 127.0000 - val_fp: 422.0000 - val_tn: 84873.0000 - val_fn: 21.0000 - val_accuracy: 0.9948 - val_precision: 0.2313 - val_recall: 0.8581 - val_auc: 0.9391 - val_sensitivity: 0.8446 - val_specificity: 0.9992
Epoch 2/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0209 - tp: 198127.0000 - fp: 1516.0000 - tn: 197504.0000 - fn: 893.0000 - accuracy: 0.9939 - precision: 0.9924 - recall: 0.9955 - auc: 0.9992 - sensitivity: 0.8827 - specificity: 0.9990 - val_loss: 0.0198 - val_tp: 127.0000 - val_fp: 378.0000 - val_tn: 84917.0000 - val_fn: 21.0000 - val_accuracy: 0.9953 - val_precision: 0.2515 - val_recall: 0.8581 - val_auc: 0.9337 - val_sensitivity: 0.8378 - val_specificity: 0.9980
Epoch 3/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0167 - tp: 198317.0000 - fp: 1201.0000 - tn: 197819.0000 - fn: 703.0000 - accuracy: 0.9952 - precision: 0.9940 - recall: 0.9965 - auc: 0.9993 - sensitivity: 0.9028 - specificity: 0.9991 - val_loss: 0.0184 - val_tp: 127.0000 - val_fp: 336.0000 - val_tn: 84959.0000 - val_fn: 21.0000 - val_accuracy: 0.9958 - val_precision: 0.2743 - val_recall: 0.8581 - val_auc: 0.9340 - val_sensitivity: 0.8378 - val_specificity: 0.9977
Epoch 4/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0142 - tp: 198490.0000 - fp: 1016.0000 - tn: 198004.0000 - fn: 530.0000 - accuracy: 0.9961 - precision: 0.9949 - recall: 0.9973 - auc: 0.9993 - sensitivity: 0.9161 - specificity: 0.9991 - val_loss: 0.0172 - val_tp: 127.0000 - val_fp: 313.0000 - val_tn: 84982.0000 - val_fn: 21.0000 - val_accuracy: 0.9961 - val_precision: 0.2886 - val_recall: 0.8581 - val_auc: 0.9342 - val_sensitivity: 0.8378 - val_specificity: 0.9983
Epoch 5/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0131 - tp: 198539.0000 - fp: 899.0000 - tn: 198121.0000 - fn: 481.0000 - accuracy: 0.9965 - precision: 0.9955 - recall: 0.9976 - auc: 0.9994 - sensitivity: 0.9196 - specificity: 0.9991 - val_loss: 0.0151 - val_tp: 126.0000 - val_fp: 262.0000 - val_tn: 85033.0000 - val_fn: 22.0000 - val_accuracy: 0.9967 - val_precision: 0.3247 - val_recall: 0.8514 - val_auc: 0.9344 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 6/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0118 - tp: 198620.0000 - fp: 812.0000 - tn: 198208.0000 - fn: 400.0000 - accuracy: 0.9970 - precision: 0.9959 - recall: 0.9980 - auc: 0.9994 - sensitivity: 0.9560 - specificity: 0.9992 - val_loss: 0.0161 - val_tp: 126.0000 - val_fp: 263.0000 - val_tn: 85032.0000 - val_fn: 22.0000 - val_accuracy: 0.9967 - val_precision: 0.3239 - val_recall: 0.8514 - val_auc: 0.9309 - val_sensitivity: 0.8378 - val_specificity: 0.9989
Epoch 7/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0111 - tp: 198672.0000 - fp: 749.0000 - tn: 198271.0000 - fn: 348.0000 - accuracy: 0.9972 - precision: 0.9962 - recall: 0.9983 - auc: 0.9994 - sensitivity: 0.9578 - specificity: 0.9992 - val_loss: 0.0151 - val_tp: 126.0000 - val_fp: 254.0000 - val_tn: 85041.0000 - val_fn: 22.0000 - val_accuracy: 0.9968 - val_precision: 0.3316 - val_recall: 0.8514 - val_auc: 0.9311 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 8/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0106 - tp: 198676.0000 - fp: 670.0000 - tn: 198350.0000 - fn: 344.0000 - accuracy: 0.9975 - precision: 0.9966 - recall: 0.9983 - auc: 0.9994 - sensitivity: 0.9607 - specificity: 0.9992 - val_loss: 0.0150 - val_tp: 126.0000 - val_fp: 251.0000 - val_tn: 85044.0000 - val_fn: 22.0000 - val_accuracy: 0.9968 - val_precision: 0.3342 - val_recall: 0.8514 - val_auc: 0.9344 - val_sensitivity: 0.8378 - val_specificity: 0.9989
Epoch 9/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0100 - tp: 198708.0000 - fp: 668.0000 - tn: 198352.0000 - fn: 312.0000 - accuracy: 0.9975 - precision: 0.9966 - recall: 0.9984 - auc: 0.9995 - sensitivity: 0.9733 - specificity: 0.9992 - val_loss: 0.0122 - val_tp: 126.0000 - val_fp: 186.0000 - val_tn: 85109.0000 - val_fn: 22.0000 - val_accuracy: 0.9976 - val_precision: 0.4038 - val_recall: 0.8514 - val_auc: 0.9313 - val_sensitivity: 0.8378 - val_specificity: 0.9989
Epoch 10/200
398040/398040 [==============================] - 4s 9us/sample - loss: 0.0093 - tp: 198743.0000 - fp: 600.0000 - tn: 198420.0000 - fn: 277.0000 - accuracy: 0.9978 - precision: 0.9970 - recall: 0.9986 - auc: 0.9995 - sensitivity: 0.9661 - specificity: 0.9993 - val_loss: 0.0142 - val_tp: 126.0000 - val_fp: 219.0000 - val_tn: 85076.0000 - val_fn: 22.0000 - val_accuracy: 0.9972 - val_precision: 0.3652 - val_recall: 0.8514 - val_auc: 0.9312 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 11/200
398040/398040 [==============================] - 4s 9us/sample - loss: 0.0088 - tp: 198754.0000 - fp: 559.0000 - tn: 198461.0000 - fn: 266.0000 - accuracy: 0.9979 - precision: 0.9972 - recall: 0.9987 - auc: 0.9995 - sensitivity: 0.9816 - specificity: 0.9993 - val_loss: 0.0126 - val_tp: 126.0000 - val_fp: 196.0000 - val_tn: 85099.0000 - val_fn: 22.0000 - val_accuracy: 0.9974 - val_precision: 0.3913 - val_recall: 0.8514 - val_auc: 0.9279 - val_sensitivity: 0.8446 - val_specificity: 0.9992
Epoch 12/200
398040/398040 [==============================] - 4s 9us/sample - loss: 0.0084 - tp: 198800.0000 - fp: 520.0000 - tn: 198500.0000 - fn: 220.0000 - accuracy: 0.9981 - precision: 0.9974 - recall: 0.9989 - auc: 0.9995 - sensitivity: 0.9820 - specificity: 0.9993 - val_loss: 0.0115 - val_tp: 126.0000 - val_fp: 166.0000 - val_tn: 85129.0000 - val_fn: 22.0000 - val_accuracy: 0.9978 - val_precision: 0.4315 - val_recall: 0.8514 - val_auc: 0.9280 - val_sensitivity: 0.8446 - val_specificity: 0.9993
Epoch 13/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198817.0000 - fp: 482.0000 - tn: 198538.0000 - fn: 203.0000 - accuracy: 0.9983 - precision: 0.9976 - recall: 0.9990 - auc: 0.9996 - sensitivity: 0.9834 - specificity: 0.9994 - val_loss: 0.0118 - val_tp: 126.0000 - val_fp: 179.0000 - val_tn: 85116.0000 - val_fn: 22.0000 - val_accuracy: 0.9976 - val_precision: 0.4131 - val_recall: 0.8514 - val_auc: 0.9281 - val_sensitivity: 0.8446 - val_specificity: 0.9988
Epoch 14/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198825.0000 - fp: 462.0000 - tn: 198558.0000 - fn: 195.0000 - accuracy: 0.9983 - precision: 0.9977 - recall: 0.9990 - auc: 0.9995 - sensitivity: 0.9902 - specificity: 0.9993 - val_loss: 0.0110 - val_tp: 126.0000 - val_fp: 146.0000 - val_tn: 85149.0000 - val_fn: 22.0000 - val_accuracy: 0.9980 - val_precision: 0.4632 - val_recall: 0.8514 - val_auc: 0.9249 - val_sensitivity: 0.8446 - val_specificity: 0.9993
Epoch 15/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198849.0000 - fp: 437.0000 - tn: 198583.0000 - fn: 171.0000 - accuracy: 0.9985 - precision: 0.9978 - recall: 0.9991 - auc: 0.9996 - sensitivity: 0.9922 - specificity: 0.9993 - val_loss: 0.0108 - val_tp: 126.0000 - val_fp: 147.0000 - val_tn: 85148.0000 - val_fn: 22.0000 - val_accuracy: 0.9980 - val_precision: 0.4615 - val_recall: 0.8514 - val_auc: 0.9249 - val_sensitivity: 0.8446 - val_specificity: 0.9992
Epoch 16/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0072 - tp: 198835.0000 - fp: 411.0000 - tn: 198609.0000 - fn: 185.0000 - accuracy: 0.9985 - precision: 0.9979 - recall: 0.9991 - auc: 0.9995 - sensitivity: 0.9888 - specificity: 0.9993 - val_loss: 0.0116 - val_tp: 126.0000 - val_fp: 164.0000 - val_tn: 85131.0000 - val_fn: 22.0000 - val_accuracy: 0.9978 - val_precision: 0.4345 - val_recall: 0.8514 - val_auc: 0.9248 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 17/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0068 - tp: 198861.0000 - fp: 401.0000 - tn: 198619.0000 - fn: 159.0000 - accuracy: 0.9986 - precision: 0.9980 - recall: 0.9992 - auc: 0.9996 - sensitivity: 0.9938 - specificity: 0.9994 - val_loss: 0.0100 - val_tp: 126.0000 - val_fp: 128.0000 - val_tn: 85167.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4961 - val_recall: 0.8514 - val_auc: 0.9250 - val_sensitivity: 0.8446 - val_specificity: 0.9993
Epoch 18/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0070 - tp: 198857.0000 - fp: 391.0000 - tn: 198629.0000 - fn: 163.0000 - accuracy: 0.9986 - precision: 0.9980 - recall: 0.9992 - auc: 0.9995 - sensitivity: 0.9890 - specificity: 0.9993 - val_loss: 0.0110 - val_tp: 126.0000 - val_fp: 145.0000 - val_tn: 85150.0000 - val_fn: 22.0000 - val_accuracy: 0.9980 - val_precision: 0.4649 - val_recall: 0.8514 - val_auc: 0.9249 - val_sensitivity: 0.8514 - val_specificity: 0.9993
Epoch 19/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0066 - tp: 198868.0000 - fp: 374.0000 - tn: 198646.0000 - fn: 152.0000 - accuracy: 0.9987 - precision: 0.9981 - recall: 0.9992 - auc: 0.9996 - sensitivity: 0.9951 - specificity: 0.9993 - val_loss: 0.0104 - val_tp: 126.0000 - val_fp: 131.0000 - val_tn: 85164.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4903 - val_recall: 0.8514 - val_auc: 0.9249 - val_sensitivity: 0.8514 - val_specificity: 0.9993
Epoch 20/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0063 - tp: 198851.0000 - fp: 356.0000 - tn: 198664.0000 - fn: 169.0000 - accuracy: 0.9987 - precision: 0.9982 - recall: 0.9992 - auc: 0.9996 - sensitivity: 0.9957 - specificity: 0.9994 - val_loss: 0.0107 - val_tp: 126.0000 - val_fp: 133.0000 - val_tn: 85162.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4865 - val_recall: 0.8514 - val_auc: 0.9283 - val_sensitivity: 0.8446 - val_specificity: 0.9993
Epoch 21/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0062 - tp: 198890.0000 - fp: 321.0000 - tn: 198699.0000 - fn: 130.0000 - accuracy: 0.9989 - precision: 0.9984 - recall: 0.9993 - auc: 0.9996 - sensitivity: 0.9966 - specificity: 0.9994 - val_loss: 0.0101 - val_tp: 126.0000 - val_fp: 121.0000 - val_tn: 85174.0000 - val_fn: 22.0000 - val_accuracy: 0.9983 - val_precision: 0.5101 - val_recall: 0.8514 - val_auc: 0.9284 - val_sensitivity: 0.8446 - val_specificity: 0.9993
Epoch 22/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0061 - tp: 198892.0000 - fp: 338.0000 - tn: 198682.0000 - fn: 128.0000 - accuracy: 0.9988 - precision: 0.9983 - recall: 0.9994 - auc: 0.9996 - sensitivity: 0.9963 - specificity: 0.9994 - val_loss: 0.0106 - val_tp: 126.0000 - val_fp: 132.0000 - val_tn: 85163.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4884 - val_recall: 0.8514 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 23/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0061 - tp: 198883.0000 - fp: 325.0000 - tn: 198695.0000 - fn: 137.0000 - accuracy: 0.9988 - precision: 0.9984 - recall: 0.9993 - auc: 0.9996 - sensitivity: 0.9967 - specificity: 0.9994 - val_loss: 0.0100 - val_tp: 126.0000 - val_fp: 118.0000 - val_tn: 85177.0000 - val_fn: 22.0000 - val_accuracy: 0.9984 - val_precision: 0.5164 - val_recall: 0.8514 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9993
Epoch 24/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0058 - tp: 198895.0000 - fp: 306.0000 - tn: 198714.0000 - fn: 125.0000 - accuracy: 0.9989 - precision: 0.9985 - recall: 0.9994 - auc: 0.9996 - sensitivity: 0.9971 - specificity: 0.9995 - val_loss: 0.0100 - val_tp: 126.0000 - val_fp: 116.0000 - val_tn: 85179.0000 - val_fn: 22.0000 - val_accuracy: 0.9984 - val_precision: 0.5207 - val_recall: 0.8514 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9993
Epoch 25/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0058 - tp: 198892.0000 - fp: 306.0000 - tn: 198714.0000 - fn: 128.0000 - accuracy: 0.9989 - precision: 0.9985 - recall: 0.9994 - auc: 0.9996 - sensitivity: 0.9968 - specificity: 0.9995 - val_loss: 0.0088 - val_tp: 126.0000 - val_fp: 105.0000 - val_tn: 85190.0000 - val_fn: 22.0000 - val_accuracy: 0.9985 - val_precision: 0.5455 - val_recall: 0.8514 - val_auc: 0.9284 - val_sensitivity: 0.8446 - val_specificity: 0.9995fn: 60.0000 - accuracy: 0.9989 - precision: 0.9985 - recall: 0.9
Epoch 26/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0056 - tp: 198902.0000 - fp: 316.0000 - tn: 198704.0000 - fn: 118.0000 - accuracy: 0.9989 - precision: 0.9984 - recall: 0.9994 - auc: 0.9996 - sensitivity: 0.9978 - specificity: 0.9994 - val_loss: 0.0091 - val_tp: 125.0000 - val_fp: 108.0000 - val_tn: 85187.0000 - val_fn: 23.0000 - val_accuracy: 0.9985 - val_precision: 0.5365 - val_recall: 0.8446 - val_auc: 0.9285 - val_sensitivity: 0.8378 - val_specificity: 0.9993
Epoch 27/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0055 - tp: 198897.0000 - fp: 303.0000 - tn: 198717.0000 - fn: 123.0000 - accuracy: 0.9989 - precision: 0.9985 - recall: 0.9994 - auc: 0.9996 - sensitivity: 0.9980 - specificity: 0.9995 - val_loss: 0.0099 - val_tp: 125.0000 - val_fp: 109.0000 - val_tn: 85186.0000 - val_fn: 23.0000 - val_accuracy: 0.9985 - val_precision: 0.5342 - val_recall: 0.8446 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 28/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0054 - tp: 198911.0000 - fp: 280.0000 - tn: 198740.0000 - fn: 109.0000 - accuracy: 0.9990 - precision: 0.9986 - recall: 0.9995 - auc: 0.9996 - sensitivity: 0.9981 - specificity: 0.9995 - val_loss: 0.0092 - val_tp: 125.0000 - val_fp: 108.0000 - val_tn: 85187.0000 - val_fn: 23.0000 - val_accuracy: 0.9985 - val_precision: 0.5365 - val_recall: 0.8446 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9994
Epoch 29/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0053 - tp: 198902.0000 - fp: 290.0000 - tn: 198730.0000 - fn: 118.0000 - accuracy: 0.9990 - precision: 0.9985 - recall: 0.9994 - auc: 0.9997 - sensitivity: 0.9984 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 125.0000 - val_fp: 108.0000 - val_tn: 85187.0000 - val_fn: 23.0000 - val_accuracy: 0.9985 - val_precision: 0.5365 - val_recall: 0.8446 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 30/200
398040/398040 [==============================] - 6s 14us/sample - loss: 0.0055 - tp: 198912.0000 - fp: 275.0000 - tn: 198745.0000 - fn: 108.0000 - accuracy: 0.9990 - precision: 0.9986 - recall: 0.9995 - auc: 0.9996 - sensitivity: 0.9982 - specificity: 0.9995 - val_loss: 0.0083 - val_tp: 125.0000 - val_fp: 88.0000 - val_tn: 85207.0000 - val_fn: 23.0000 - val_accuracy: 0.9987 - val_precision: 0.5869 - val_recall: 0.8446 - val_auc: 0.9285 - val_sensitivity: 0.8446 - val_specificity: 0.9994
Epoch 31/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0054 - tp: 198908.0000 - fp: 260.0000 - tn: 198760.0000 - fn: 112.0000 - accuracy: 0.9991 - precision: 0.9987 - recall: 0.9994 - auc: 0.9996 - sensitivity: 0.9985 - specificity: 0.9995 - val_loss: 0.0083 - val_tp: 125.0000 - val_fp: 94.0000 - val_tn: 85201.0000 - val_fn: 23.0000 - val_accuracy: 0.9986 - val_precision: 0.5708 - val_recall: 0.8446 - val_auc: 0.9285 - val_sensitivity: 0.8446 - val_specificity: 0.9995
Epoch 32/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0048 - tp: 198929.0000 - fp: 248.0000 - tn: 198772.0000 - fn: 91.0000 - accuracy: 0.9991 - precision: 0.9988 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9992 - specificity: 0.9995 - val_loss: 0.0095 - val_tp: 125.0000 - val_fp: 106.0000 - val_tn: 85189.0000 - val_fn: 23.0000 - val_accuracy: 0.9985 - val_precision: 0.5411 - val_recall: 0.8446 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 33/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0049 - tp: 198909.0000 - fp: 239.0000 - tn: 198781.0000 - fn: 111.0000 - accuracy: 0.9991 - precision: 0.9988 - recall: 0.9994 - auc: 0.9997 - sensitivity: 0.9991 - specificity: 0.9995 - val_loss: 0.0088 - val_tp: 125.0000 - val_fp: 99.0000 - val_tn: 85196.0000 - val_fn: 23.0000 - val_accuracy: 0.9986 - val_precision: 0.5580 - val_recall: 0.8446 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 34/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0050 - tp: 198905.0000 - fp: 241.0000 - tn: 198779.0000 - fn: 115.0000 - accuracy: 0.9991 - precision: 0.9988 - recall: 0.9994 - auc: 0.9997 - sensitivity: 0.9990 - specificity: 0.9995 - val_loss: 0.0082 - val_tp: 123.0000 - val_fp: 83.0000 - val_tn: 85212.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5971 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 35/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0049 - tp: 198922.0000 - fp: 239.0000 - tn: 198781.0000 - fn: 98.0000 - accuracy: 0.9992 - precision: 0.9988 - recall: 0.9995 - auc: 0.9996 - sensitivity: 0.9993 - specificity: 0.9995 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 69.0000 - val_tn: 85226.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6406 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 36/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0050 - tp: 198908.0000 - fp: 239.0000 - tn: 198781.0000 - fn: 112.0000 - accuracy: 0.9991 - precision: 0.9988 - recall: 0.9994 - auc: 0.9997 - sensitivity: 0.9990 - specificity: 0.9995 - val_loss: 0.0082 - val_tp: 123.0000 - val_fp: 84.0000 - val_tn: 85211.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5942 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 37/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0048 - tp: 198914.0000 - fp: 233.0000 - tn: 198787.0000 - fn: 106.0000 - accuracy: 0.9991 - precision: 0.9988 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9993 - specificity: 0.9996 - val_loss: 0.0081 - val_tp: 123.0000 - val_fp: 79.0000 - val_tn: 85216.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6089 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 38/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0046 - tp: 198913.0000 - fp: 205.0000 - tn: 198815.0000 - fn: 107.0000 - accuracy: 0.9992 - precision: 0.9990 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9994 - specificity: 0.9996 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 84.0000 - val_tn: 85211.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5942 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 39/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0046 - tp: 198932.0000 - fp: 221.0000 - tn: 198799.0000 - fn: 88.0000 - accuracy: 0.9992 - precision: 0.9989 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9995 - specificity: 0.9996 - val_loss: 0.0080 - val_tp: 123.0000 - val_fp: 75.0000 - val_tn: 85220.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6212 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 40/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0045 - tp: 198931.0000 - fp: 227.0000 - tn: 198793.0000 - fn: 89.0000 - accuracy: 0.9992 - precision: 0.9989 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9994 - specificity: 0.9996 - val_loss: 0.0079 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 41/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0045 - tp: 198926.0000 - fp: 202.0000 - tn: 198818.0000 - fn: 94.0000 - accuracy: 0.9993 - precision: 0.9990 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9995 - specificity: 0.9995 - val_loss: 0.0078 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 42/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0046 - tp: 198919.0000 - fp: 205.0000 - tn: 198815.0000 - fn: 101.0000 - accuracy: 0.9992 - precision: 0.9990 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9995 - specificity: 0.9996 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 83.0000 - val_tn: 85212.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5971 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 43/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0044 - tp: 198934.0000 - fp: 192.0000 - tn: 198828.0000 - fn: 86.0000 - accuracy: 0.9993 - precision: 0.9990 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9995 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 62.0000 - val_tn: 85233.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6649 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 44/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0046 - tp: 198926.0000 - fp: 210.0000 - tn: 198810.0000 - fn: 94.0000 - accuracy: 0.9992 - precision: 0.9989 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9995 - specificity: 0.9995 - val_loss: 0.0079 - val_tp: 123.0000 - val_fp: 71.0000 - val_tn: 85224.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6340 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 45/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0043 - tp: 198928.0000 - fp: 190.0000 - tn: 198830.0000 - fn: 92.0000 - accuracy: 0.9993 - precision: 0.9990 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 46/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0044 - tp: 198919.0000 - fp: 210.0000 - tn: 198810.0000 - fn: 101.0000 - accuracy: 0.9992 - precision: 0.9989 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9995 - specificity: 0.9996 - val_loss: 0.0077 - val_tp: 123.0000 - val_fp: 66.0000 - val_tn: 85229.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6508 - val_recall: 0.8311 - val_auc: 0.9286 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 47/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0042 - tp: 198941.0000 - fp: 184.0000 - tn: 198836.0000 - fn: 79.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0077 - val_tp: 123.0000 - val_fp: 66.0000 - val_tn: 85229.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6508 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 48/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0041 - tp: 198931.0000 - fp: 176.0000 - tn: 198844.0000 - fn: 89.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 61.0000 - val_tn: 85234.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6685 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 49/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0042 - tp: 198941.0000 - fp: 188.0000 - tn: 198832.0000 - fn: 79.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9996 - val_loss: 0.0075 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 50/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0041 - tp: 198926.0000 - fp: 171.0000 - tn: 198849.0000 - fn: 94.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 63.0000 - val_tn: 85232.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6613 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 51/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0042 - tp: 198934.0000 - fp: 187.0000 - tn: 198833.0000 - fn: 86.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 60.0000 - val_tn: 85235.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6721 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 52/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0040 - tp: 198946.0000 - fp: 178.0000 - tn: 198842.0000 - fn: 74.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0083 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 53/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0043 - tp: 198930.0000 - fp: 180.0000 - tn: 198840.0000 - fn: 90.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 62.0000 - val_tn: 85233.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6649 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 54/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0040 - tp: 198946.0000 - fp: 168.0000 - tn: 198852.0000 - fn: 74.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0078 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9286 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 55/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0041 - tp: 198932.0000 - fp: 188.0000 - tn: 198832.0000 - fn: 88.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9996 - val_loss: 0.0075 - val_tp: 123.0000 - val_fp: 61.0000 - val_tn: 85234.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6685 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 56/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0039 - tp: 198934.0000 - fp: 168.0000 - tn: 198852.0000 - fn: 86.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0079 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 57/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0039 - tp: 198940.0000 - fp: 163.0000 - tn: 198857.0000 - fn: 80.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 62.0000 - val_tn: 85233.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6649 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 58/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0039 - tp: 198940.0000 - fp: 181.0000 - tn: 198839.0000 - fn: 80.0000 - accuracy: 0.9993 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 59.0000 - val_tn: 85236.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6758 - val_recall: 0.8311 - val_auc: 0.9253 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 59/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0040 - tp: 198942.0000 - fp: 174.0000 - tn: 198846.0000 - fn: 78.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 55.0000 - val_tn: 85240.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6910 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 60/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0040 - tp: 198933.0000 - fp: 164.0000 - tn: 198856.0000 - fn: 87.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9996 - specificity: 0.9996 - val_loss: 0.0078 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 61/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0039 - tp: 198936.0000 - fp: 165.0000 - tn: 198855.0000 - fn: 84.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 60.0000 - val_tn: 85235.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6721 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 62/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198940.0000 - fp: 155.0000 - tn: 198865.0000 - fn: 80.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 55.0000 - val_tn: 85240.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6910 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 63/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0038 - tp: 198945.0000 - fp: 155.0000 - tn: 198865.0000 - fn: 75.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 64/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198953.0000 - fp: 157.0000 - tn: 198863.0000 - fn: 67.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0078 - val_tp: 123.0000 - val_fp: 67.0000 - val_tn: 85228.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6474 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 65/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198939.0000 - fp: 153.0000 - tn: 198867.0000 - fn: 81.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0079 - val_tp: 123.0000 - val_fp: 65.0000 - val_tn: 85230.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6543 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8378 - val_specificity: 0.9994
Epoch 66/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0038 - tp: 198950.0000 - fp: 164.0000 - tn: 198856.0000 - fn: 70.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 53.0000 - val_tn: 85242.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6989 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 67/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198941.0000 - fp: 145.0000 - tn: 198875.0000 - fn: 79.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0076 - val_tp: 123.0000 - val_fp: 62.0000 - val_tn: 85233.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6649 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 68/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198948.0000 - fp: 151.0000 - tn: 198869.0000 - fn: 72.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 56.0000 - val_tn: 85239.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6872 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 69/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0039 - tp: 198934.0000 - fp: 154.0000 - tn: 198866.0000 - fn: 86.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 59.0000 - val_tn: 85236.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6758 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 70/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0038 - tp: 198952.0000 - fp: 152.0000 - tn: 198868.0000 - fn: 68.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 71/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198944.0000 - fp: 144.0000 - tn: 198876.0000 - fn: 76.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0068 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9996
Epoch 72/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198929.0000 - fp: 147.0000 - tn: 198873.0000 - fn: 91.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9995 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 53.0000 - val_tn: 85242.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6989 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 73/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198939.0000 - fp: 149.0000 - tn: 198871.0000 - fn: 81.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 56.0000 - val_tn: 85239.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6872 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 74/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0038 - tp: 198941.0000 - fp: 155.0000 - tn: 198865.0000 - fn: 79.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 53.0000 - val_tn: 85242.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6989 - val_recall: 0.8311 - val_auc: 0.9253 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 75/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198957.0000 - fp: 148.0000 - tn: 198872.0000 - fn: 63.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 55.0000 - val_tn: 85240.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6910 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 76/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0036 - tp: 198946.0000 - fp: 147.0000 - tn: 198873.0000 - fn: 74.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7069 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 77/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0037 - tp: 198947.0000 - fp: 147.0000 - tn: 198873.0000 - fn: 73.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 78/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0036 - tp: 198948.0000 - fp: 140.0000 - tn: 198880.0000 - fn: 72.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 79/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0037 - tp: 198938.0000 - fp: 144.0000 - tn: 198876.0000 - fn: 82.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 80/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0037 - tp: 198947.0000 - fp: 146.0000 - tn: 198874.0000 - fn: 73.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 49.0000 - val_tn: 85246.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7151 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 81/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0037 - tp: 198940.0000 - fp: 148.0000 - tn: 198872.0000 - fn: 80.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9253 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 82/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198936.0000 - fp: 141.0000 - tn: 198879.0000 - fn: 84.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 83/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0036 - tp: 198940.0000 - fp: 151.0000 - tn: 198869.0000 - fn: 80.0000 - accuracy: 0.9994 - precision: 0.9992 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 84/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198941.0000 - fp: 149.0000 - tn: 198871.0000 - fn: 79.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 52.0000 - val_tn: 85243.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7029 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 85/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0036 - tp: 198943.0000 - fp: 130.0000 - tn: 198890.0000 - fn: 77.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 52.0000 - val_tn: 85243.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7029 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 86/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0037 - tp: 198940.0000 - fp: 143.0000 - tn: 198877.0000 - fn: 80.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 49.0000 - val_tn: 85246.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7151 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 87/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198941.0000 - fp: 144.0000 - tn: 198876.0000 - fn: 79.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 88/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0037 - tp: 198952.0000 - fp: 149.0000 - tn: 198871.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0075 - val_tp: 123.0000 - val_fp: 56.0000 - val_tn: 85239.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6872 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 89/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198947.0000 - fp: 144.0000 - tn: 198876.0000 - fn: 73.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0074 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 90/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0035 - tp: 198939.0000 - fp: 142.0000 - tn: 198878.0000 - fn: 81.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7069 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 91/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0035 - tp: 198939.0000 - fp: 139.0000 - tn: 198881.0000 - fn: 81.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 92/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0034 - tp: 198955.0000 - fp: 124.0000 - tn: 198896.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 93/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198943.0000 - fp: 138.0000 - tn: 198882.0000 - fn: 77.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7069 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 94/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198952.0000 - fp: 139.0000 - tn: 198881.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 49.0000 - val_tn: 85246.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7151 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 95/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198946.0000 - fp: 123.0000 - tn: 198897.0000 - fn: 74.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7069 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 96/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0036 - tp: 198945.0000 - fp: 134.0000 - tn: 198886.0000 - fn: 75.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7069 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 97/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198953.0000 - fp: 138.0000 - tn: 198882.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0067 - val_tp: 123.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7410 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9995
Epoch 98/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198950.0000 - fp: 128.0000 - tn: 198892.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 52.0000 - val_tn: 85243.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7029 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 99/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0036 - tp: 198942.0000 - fp: 136.0000 - tn: 198884.0000 - fn: 78.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 54.0000 - val_tn: 85241.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.6949 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 100/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198952.0000 - fp: 137.0000 - tn: 198883.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 123.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7235 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 101/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198949.0000 - fp: 126.0000 - tn: 198894.0000 - fn: 71.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 102/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198955.0000 - fp: 135.0000 - tn: 198885.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7176 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 103/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198950.0000 - fp: 132.0000 - tn: 198888.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7069 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 104/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198953.0000 - fp: 130.0000 - tn: 198890.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7262 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 105/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198952.0000 - fp: 121.0000 - tn: 198899.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9996
Epoch 106/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198943.0000 - fp: 144.0000 - tn: 198876.0000 - fn: 77.0000 - accuracy: 0.9994 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7193 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 107/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198952.0000 - fp: 124.0000 - tn: 198896.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0071 - val_tp: 123.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7193 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 108/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0036 - tp: 198950.0000 - fp: 118.0000 - tn: 198902.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0068 - val_tp: 123.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7235 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 109/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198951.0000 - fp: 133.0000 - tn: 198887.0000 - fn: 69.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 52.0000 - val_tn: 85243.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7029 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 110/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198953.0000 - fp: 126.0000 - tn: 198894.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0073 - val_tp: 123.0000 - val_fp: 52.0000 - val_tn: 85243.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7029 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 111/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198953.0000 - fp: 129.0000 - tn: 198891.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 112/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198956.0000 - fp: 132.0000 - tn: 198888.0000 - fn: 64.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 123.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7410 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 113/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0036 - tp: 198948.0000 - fp: 126.0000 - tn: 198894.0000 - fn: 72.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0067 - val_tp: 123.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7278 - val_recall: 0.8311 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 114/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0035 - tp: 198946.0000 - fp: 124.0000 - tn: 198896.0000 - fn: 74.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7093 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 115/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198949.0000 - fp: 133.0000 - tn: 198887.0000 - fn: 71.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7219 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9993
Epoch 116/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198953.0000 - fp: 136.0000 - tn: 198884.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7262 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 117/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198949.0000 - fp: 129.0000 - tn: 198891.0000 - fn: 71.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0066 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 118/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198944.0000 - fp: 132.0000 - tn: 198888.0000 - fn: 76.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0069 - val_tp: 123.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7235 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 119/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0035 - tp: 198954.0000 - fp: 124.0000 - tn: 198896.0000 - fn: 66.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0069 - val_tp: 123.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7235 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 120/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198952.0000 - fp: 132.0000 - tn: 198888.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 123.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7278 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 121/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198950.0000 - fp: 115.0000 - tn: 198905.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 122/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198956.0000 - fp: 120.0000 - tn: 198900.0000 - fn: 64.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 123/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198964.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 56.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 124/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0035 - tp: 198944.0000 - fp: 130.0000 - tn: 198890.0000 - fn: 76.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9996
Epoch 125/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 122.0000 - tn: 198898.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 126/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198956.0000 - fp: 117.0000 - tn: 198903.0000 - fn: 64.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995fp: 88.0000 - tn: 147772.0000 - fn: 42.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificit - ETA: 0s - loss: 0.0033 - tp: 173527.0000 - fp: 99.0000 - tn: 173457.0000 - fn: 53.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - spe
Epoch 127/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198948.0000 - fp: 128.0000 - tn: 198892.0000 - fn: 72.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0068 - val_tp: 123.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 25.0000 - val_accuracy: 0.9992 - val_precision: 0.7235 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 128/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198957.0000 - fp: 120.0000 - tn: 198900.0000 - fn: 63.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 129/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198952.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 121.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 27.0000 - val_accuracy: 0.9992 - val_precision: 0.7289 - val_recall: 0.8176 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 130/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198955.0000 - fp: 121.0000 - tn: 198899.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 131/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198948.0000 - fp: 117.0000 - tn: 198903.0000 - fn: 72.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 132/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198947.0000 - fp: 127.0000 - tn: 198893.0000 - fn: 73.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0066 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9995
Epoch 133/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198950.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0072 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 134/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198962.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 58.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 121.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 27.0000 - val_accuracy: 0.9992 - val_precision: 0.7333 - val_recall: 0.8176 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 135/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198949.0000 - fp: 115.0000 - tn: 198905.0000 - fn: 71.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7176 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9994
Epoch 136/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198951.0000 - fp: 124.0000 - tn: 198896.0000 - fn: 69.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7262 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 137/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198948.0000 - fp: 123.0000 - tn: 198897.0000 - fn: 72.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 123.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 25.0000 - val_accuracy: 0.9991 - val_precision: 0.7110 - val_recall: 0.8311 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 138/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198951.0000 - fp: 127.0000 - tn: 198893.0000 - fn: 69.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7219 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 139/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198947.0000 - fp: 125.0000 - tn: 198895.0000 - fn: 73.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 50.0000 - val_tn: 85245.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7093 - val_recall: 0.8243 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 140/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0034 - tp: 198957.0000 - fp: 118.0000 - tn: 198902.0000 - fn: 63.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 51.0000 - val_tn: 85244.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7052 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 141/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0031 - tp: 198954.0000 - fp: 111.0000 - tn: 198909.0000 - fn: 66.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0071 - val_tp: 122.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7176 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9995
Epoch 142/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0034 - tp: 198950.0000 - fp: 124.0000 - tn: 198896.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9996 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 143/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 109.0000 - tn: 198911.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7219 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 144/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198955.0000 - fp: 132.0000 - tn: 198888.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 145/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198956.0000 - fp: 106.0000 - tn: 198914.0000 - fn: 64.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 146/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198958.0000 - fp: 119.0000 - tn: 198901.0000 - fn: 62.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0066 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 147/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198946.0000 - fp: 121.0000 - tn: 198899.0000 - fn: 74.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 148/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 115.0000 - tn: 198905.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 149/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198952.0000 - fp: 113.0000 - tn: 198907.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9996
Epoch 150/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0031 - tp: 198958.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 62.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9993
Epoch 151/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0035 - tp: 198951.0000 - fp: 130.0000 - tn: 198890.0000 - fn: 69.0000 - accuracy: 0.9995 - precision: 0.9993 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 152/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198956.0000 - fp: 106.0000 - tn: 198914.0000 - fn: 64.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9254 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 153/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198950.0000 - fp: 123.0000 - tn: 198897.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9996
Epoch 154/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198957.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 63.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 155/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198955.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 156/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 110.0000 - tn: 198910.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0075 - val_tp: 122.0000 - val_fp: 56.0000 - val_tn: 85239.0000 - val_fn: 26.0000 - val_accuracy: 0.9990 - val_precision: 0.6854 - val_recall: 0.8243 - val_auc: 0.9219 - val_sensitivity: 0.8311 - val_specificity: 0.9993
Epoch 157/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198958.0000 - fp: 115.0000 - tn: 198905.0000 - fn: 62.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9152 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 158/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0032 - tp: 198953.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 159/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198948.0000 - fp: 120.0000 - tn: 198900.0000 - fn: 72.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 160/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0034 - tp: 198953.0000 - fp: 121.0000 - tn: 198899.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 161/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198950.0000 - fp: 116.0000 - tn: 198904.0000 - fn: 70.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 53.0000 - val_tn: 85242.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.6971 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 162/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198957.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 63.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 163/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 118.0000 - tn: 198902.0000 - fn: 67.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 164/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0032 - tp: 198959.0000 - fp: 106.0000 - tn: 198914.0000 - fn: 61.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0065 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 165/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198953.0000 - fp: 111.0000 - tn: 198909.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 166/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198959.0000 - fp: 114.0000 - tn: 198906.0000 - fn: 61.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 167/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0031 - tp: 198965.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 55.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 168/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 105.0000 - tn: 198915.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9996
Epoch 169/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198957.0000 - fp: 111.0000 - tn: 198909.0000 - fn: 63.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8446 - val_specificity: 0.9996
Epoch 170/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198954.0000 - fp: 114.0000 - tn: 198906.0000 - fn: 66.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 171/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198954.0000 - fp: 114.0000 - tn: 198906.0000 - fn: 66.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7262 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9993
Epoch 172/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0031 - tp: 198955.0000 - fp: 115.0000 - tn: 198905.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 173/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198963.0000 - fp: 109.0000 - tn: 198911.0000 - fn: 57.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 174/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198955.0000 - fp: 119.0000 - tn: 198901.0000 - fn: 65.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 175/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198946.0000 - fp: 123.0000 - tn: 198897.0000 - fn: 74.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 176/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0031 - tp: 198957.0000 - fp: 109.0000 - tn: 198911.0000 - fn: 63.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 177/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198958.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 62.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 178/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0033 - tp: 198947.0000 - fp: 111.0000 - tn: 198909.0000 - fn: 73.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 179/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198958.0000 - fp: 119.0000 - tn: 198901.0000 - fn: 62.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0066 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 180/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0030 - tp: 198958.0000 - fp: 105.0000 - tn: 198915.0000 - fn: 62.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9993
Epoch 181/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198962.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 58.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0066 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 182/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0033 - tp: 198953.0000 - fp: 110.0000 - tn: 198910.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9992
Epoch 183/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0032 - tp: 198954.0000 - fp: 111.0000 - tn: 198909.0000 - fn: 66.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 184/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0032 - tp: 198960.0000 - fp: 114.0000 - tn: 198906.0000 - fn: 60.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 40.0000 - val_tn: 85255.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7531 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 185/200
398040/398040 [==============================] - 6s 15us/sample - loss: 0.0031 - tp: 198953.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 67.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.99930000 - tn: 170383.0000 - fn: 60.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9996 - auc: 0.9997 - sensitivity: 0.9998
Epoch 186/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0032 - tp: 198960.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 60.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9152 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 187/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198956.0000 - fp: 113.0000 - tn: 198907.0000 - fn: 64.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9152 - val_sensitivity: 0.8311 - val_specificity: 0.9993
Epoch 188/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0030 - tp: 198955.0000 - fp: 102.0000 - tn: 198918.0000 - fn: 65.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9152 - val_sensitivity: 0.8311 - val_specificity: 0.9993
Epoch 189/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0030 - tp: 198962.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 58.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0070 - val_tp: 122.0000 - val_fp: 46.0000 - val_tn: 85249.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7262 - val_recall: 0.8243 - val_auc: 0.9152 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 190/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198952.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 68.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 191/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198954.0000 - fp: 119.0000 - tn: 198901.0000 - fn: 66.0000 - accuracy: 0.9995 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8378 - val_specificity: 0.9996
Epoch 192/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198954.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 66.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 43.0000 - val_tn: 85252.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7394 - val_recall: 0.8243 - val_auc: 0.9186 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 193/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0030 - tp: 198955.0000 - fp: 108.0000 - tn: 198912.0000 - fn: 65.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9187 - val_sensitivity: 0.8378 - val_specificity: 0.9995
Epoch 194/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0030 - tp: 198959.0000 - fp: 102.0000 - tn: 198918.0000 - fn: 61.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9998 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 42.0000 - val_tn: 85253.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7439 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 195/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0031 - tp: 198966.0000 - fp: 106.0000 - tn: 198914.0000 - fn: 54.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 41.0000 - val_tn: 85254.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7485 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 196/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0032 - tp: 198954.0000 - fp: 106.0000 - tn: 198914.0000 - fn: 66.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0067 - val_tp: 122.0000 - val_fp: 45.0000 - val_tn: 85250.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7305 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 197/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198959.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 61.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0066 - val_tp: 122.0000 - val_fp: 44.0000 - val_tn: 85251.0000 - val_fn: 26.0000 - val_accuracy: 0.9992 - val_precision: 0.7349 - val_recall: 0.8243 - val_auc: 0.9153 - val_sensitivity: 0.8311 - val_specificity: 0.9995
Epoch 198/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198958.0000 - fp: 104.0000 - tn: 198916.0000 - fn: 62.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0068 - val_tp: 122.0000 - val_fp: 47.0000 - val_tn: 85248.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7219 - val_recall: 0.8243 - val_auc: 0.9152 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 199/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0032 - tp: 198960.0000 - fp: 112.0000 - tn: 198908.0000 - fn: 60.0000 - accuracy: 0.9996 - precision: 0.9994 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7176 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9996
Epoch 200/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0031 - tp: 198957.0000 - fp: 103.0000 - tn: 198917.0000 - fn: 63.0000 - accuracy: 0.9996 - precision: 0.9995 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9997 - val_loss: 0.0069 - val_tp: 122.0000 - val_fp: 48.0000 - val_tn: 85247.0000 - val_fn: 26.0000 - val_accuracy: 0.9991 - val_precision: 0.7176 - val_recall: 0.8243 - val_auc: 0.9220 - val_sensitivity: 0.8311 - val_specificity: 0.9995

#cost_sensitive weights to punish to the false negatives
run_5_weight_history = model_5_weight.fit(x=trainX_scaled,y=trainy_scaled,class_weight={0:1,1:5},validation_data=(testX,testy),batch_size=BATCH_SIZE,epochs=EPOCHS,callbacks=[es])

Train on 398040 samples, validate on 85443 samples
Epoch 1/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.3123 - tp: 195595.0000 - fp: 34194.0000 - tn: 164826.0000 - fn: 3425.0000 - accuracy: 0.9055 - precision: 0.8512 - recall: 0.9828 - auc: 0.9866 - sensitivity: 0.5200 - specificity: 0.9915 - val_loss: 0.0457 - val_tp: 128.0000 - val_fp: 956.0000 - val_tn: 84339.0000 - val_fn: 20.0000 - val_accuracy: 0.9886 - val_precision: 0.1181 - val_recall: 0.8649 - val_auc: 0.9575 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9954
Epoch 2/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0424 - tp: 198755.0000 - fp: 3195.0000 - tn: 195825.0000 - fn: 265.0000 - accuracy: 0.9913 - precision: 0.9842 - recall: 0.9987 - auc: 0.9986 - sensitivity: 0.0000e+00 - specificity: 0.9978 - val_loss: 0.0342 - val_tp: 127.0000 - val_fp: 610.0000 - val_tn: 84685.0000 - val_fn: 21.0000 - val_accuracy: 0.9926 - val_precision: 0.1723 - val_recall: 0.8581 - val_auc: 0.9389 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9972
Epoch 3/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0309 - tp: 198816.0000 - fp: 2038.0000 - tn: 196982.0000 - fn: 204.0000 - accuracy: 0.9944 - precision: 0.9899 - recall: 0.9990 - auc: 0.9987 - sensitivity: 0.0000e+00 - specificity: 0.9979 - val_loss: 0.0305 - val_tp: 127.0000 - val_fp: 515.0000 - val_tn: 84780.0000 - val_fn: 21.0000 - val_accuracy: 0.9937 - val_precision: 0.1978 - val_recall: 0.8581 - val_auc: 0.9331 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9975
Epoch 4/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0269 - tp: 198854.0000 - fp: 1719.0000 - tn: 197301.0000 - fn: 166.0000 - accuracy: 0.9953 - precision: 0.9914 - recall: 0.9992 - auc: 0.9987 - sensitivity: 0.0000e+00 - specificity: 0.9978 - val_loss: 0.0294 - val_tp: 127.0000 - val_fp: 489.0000 - val_tn: 84806.0000 - val_fn: 21.0000 - val_accuracy: 0.9940 - val_precision: 0.2062 - val_recall: 0.8581 - val_auc: 0.9333 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9975
Epoch 5/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0244 - tp: 198885.0000 - fp: 1515.0000 - tn: 197505.0000 - fn: 135.0000 - accuracy: 0.9959 - precision: 0.9924 - recall: 0.9993 - auc: 0.9988 - sensitivity: 0.0000e+00 - specificity: 0.9979 - val_loss: 0.0282 - val_tp: 127.0000 - val_fp: 446.0000 - val_tn: 84849.0000 - val_fn: 21.0000 - val_accuracy: 0.9945 - val_precision: 0.2216 - val_recall: 0.8581 - val_auc: 0.9335 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9976
Epoch 6/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0233 - tp: 198885.0000 - fp: 1451.0000 - tn: 197569.0000 - fn: 135.0000 - accuracy: 0.9960 - precision: 0.9928 - recall: 0.9993 - auc: 0.9989 - sensitivity: 0.9687 - specificity: 0.9981 - val_loss: 0.0255 - val_tp: 127.0000 - val_fp: 401.0000 - val_tn: 84894.0000 - val_fn: 21.0000 - val_accuracy: 0.9951 - val_precision: 0.2405 - val_recall: 0.8581 - val_auc: 0.9338 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9974
Epoch 7/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0217 - tp: 198909.0000 - fp: 1297.0000 - tn: 197723.0000 - fn: 111.0000 - accuracy: 0.9965 - precision: 0.9935 - recall: 0.9994 - auc: 0.9989 - sensitivity: 0.9714 - specificity: 0.9981 - val_loss: 0.0284 - val_tp: 127.0000 - val_fp: 453.0000 - val_tn: 84842.0000 - val_fn: 21.0000 - val_accuracy: 0.9945 - val_precision: 0.2190 - val_recall: 0.8581 - val_auc: 0.9302 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9975
Epoch 8/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0210 - tp: 198904.0000 - fp: 1235.0000 - tn: 197785.0000 - fn: 116.0000 - accuracy: 0.9966 - precision: 0.9938 - recall: 0.9994 - auc: 0.9990 - sensitivity: 0.9722 - specificity: 0.9982 - val_loss: 0.0242 - val_tp: 126.0000 - val_fp: 372.0000 - val_tn: 84923.0000 - val_fn: 22.0000 - val_accuracy: 0.9954 - val_precision: 0.2530 - val_recall: 0.8514 - val_auc: 0.9338 - val_sensitivity: 0.8446 - val_specificity: 0.9971
Epoch 9/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0200 - tp: 198914.0000 - fp: 1200.0000 - tn: 197820.0000 - fn: 106.0000 - accuracy: 0.9967 - precision: 0.9940 - recall: 0.9995 - auc: 0.9990 - sensitivity: 0.9744 - specificity: 0.9983 - val_loss: 0.0216 - val_tp: 126.0000 - val_fp: 326.0000 - val_tn: 84969.0000 - val_fn: 22.0000 - val_accuracy: 0.9959 - val_precision: 0.2788 - val_recall: 0.8514 - val_auc: 0.9305 - val_sensitivity: 0.8446 - val_specificity: 0.9971
Epoch 10/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0190 - tp: 198932.0000 - fp: 1118.0000 - tn: 197902.0000 - fn: 88.0000 - accuracy: 0.9970 - precision: 0.9944 - recall: 0.9996 - auc: 0.9991 - sensitivity: 0.9760 - specificity: 0.9984 - val_loss: 0.0209 - val_tp: 126.0000 - val_fp: 327.0000 - val_tn: 84968.0000 - val_fn: 22.0000 - val_accuracy: 0.9959 - val_precision: 0.2781 - val_recall: 0.8514 - val_auc: 0.9304 - val_sensitivity: 0.8446 - val_specificity: 0.9967
Epoch 11/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0181 - tp: 198932.0000 - fp: 1043.0000 - tn: 197977.0000 - fn: 88.0000 - accuracy: 0.9972 - precision: 0.9948 - recall: 0.9996 - auc: 0.9991 - sensitivity: 0.9789 - specificity: 0.9984 - val_loss: 0.0220 - val_tp: 126.0000 - val_fp: 327.0000 - val_tn: 84968.0000 - val_fn: 22.0000 - val_accuracy: 0.9959 - val_precision: 0.2781 - val_recall: 0.8514 - val_auc: 0.9304 - val_sensitivity: 0.8446 - val_specificity: 0.9969
Epoch 12/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0174 - tp: 198942.0000 - fp: 968.0000 - tn: 198052.0000 - fn: 78.0000 - accuracy: 0.9974 - precision: 0.9952 - recall: 0.9996 - auc: 0.9992 - sensitivity: 0.9798 - specificity: 0.9985 - val_loss: 0.0211 - val_tp: 126.0000 - val_fp: 318.0000 - val_tn: 84977.0000 - val_fn: 22.0000 - val_accuracy: 0.9960 - val_precision: 0.2838 - val_recall: 0.8514 - val_auc: 0.9304 - val_sensitivity: 0.8446 - val_specificity: 0.9971
Epoch 13/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0164 - tp: 198956.0000 - fp: 902.0000 - tn: 198118.0000 - fn: 64.0000 - accuracy: 0.9976 - precision: 0.9955 - recall: 0.9997 - auc: 0.9992 - sensitivity: 0.9797 - specificity: 0.9986 - val_loss: 0.0216 - val_tp: 126.0000 - val_fp: 311.0000 - val_tn: 84984.0000 - val_fn: 22.0000 - val_accuracy: 0.9961 - val_precision: 0.2883 - val_recall: 0.8514 - val_auc: 0.9304 - val_sensitivity: 0.8446 - val_specificity: 0.9969
Epoch 14/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0159 - tp: 198945.0000 - fp: 856.0000 - tn: 198164.0000 - fn: 75.0000 - accuracy: 0.9977 - precision: 0.9957 - recall: 0.9996 - auc: 0.9993 - sensitivity: 0.9832 - specificity: 0.9987 - val_loss: 0.0184 - val_tp: 126.0000 - val_fp: 259.0000 - val_tn: 85036.0000 - val_fn: 22.0000 - val_accuracy: 0.9967 - val_precision: 0.3273 - val_recall: 0.8514 - val_auc: 0.9305 - val_sensitivity: 0.8446 - val_specificity: 0.9970
Epoch 15/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0155 - tp: 198940.0000 - fp: 789.0000 - tn: 198231.0000 - fn: 80.0000 - accuracy: 0.9978 - precision: 0.9960 - recall: 0.9996 - auc: 0.9992 - sensitivity: 0.9832 - specificity: 0.9987 - val_loss: 0.0199 - val_tp: 126.0000 - val_fp: 293.0000 - val_tn: 85002.0000 - val_fn: 22.0000 - val_accuracy: 0.9963 - val_precision: 0.3007 - val_recall: 0.8514 - val_auc: 0.9305 - val_sensitivity: 0.8446 - val_specificity: 0.9972
Epoch 16/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0151 - tp: 198956.0000 - fp: 737.0000 - tn: 198283.0000 - fn: 64.0000 - accuracy: 0.9980 - precision: 0.9963 - recall: 0.9997 - auc: 0.9993 - sensitivity: 0.9847 - specificity: 0.9987 - val_loss: 0.0180 - val_tp: 126.0000 - val_fp: 263.0000 - val_tn: 85032.0000 - val_fn: 22.0000 - val_accuracy: 0.9967 - val_precision: 0.3239 - val_recall: 0.8514 - val_auc: 0.9275 - val_sensitivity: 0.8446 - val_specificity: 0.9973
Epoch 17/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0145 - tp: 198949.0000 - fp: 682.0000 - tn: 198338.0000 - fn: 71.0000 - accuracy: 0.9981 - precision: 0.9966 - recall: 0.9996 - auc: 0.9993 - sensitivity: 0.9853 - specificity: 0.9988 - val_loss: 0.0179 - val_tp: 126.0000 - val_fp: 262.0000 - val_tn: 85033.0000 - val_fn: 22.0000 - val_accuracy: 0.9967 - val_precision: 0.3247 - val_recall: 0.8514 - val_auc: 0.9307 - val_sensitivity: 0.8446 - val_specificity: 0.9972
Epoch 18/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0141 - tp: 198951.0000 - fp: 644.0000 - tn: 198376.0000 - fn: 69.0000 - accuracy: 0.9982 - precision: 0.9968 - recall: 0.9997 - auc: 0.9993 - sensitivity: 0.9860 - specificity: 0.9989 - val_loss: 0.0160 - val_tp: 125.0000 - val_fp: 213.0000 - val_tn: 85082.0000 - val_fn: 23.0000 - val_accuracy: 0.9972 - val_precision: 0.3698 - val_recall: 0.8446 - val_auc: 0.9308 - val_sensitivity: 0.8446 - val_specificity: 0.9966
Epoch 19/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0138 - tp: 198959.0000 - fp: 625.0000 - tn: 198395.0000 - fn: 61.0000 - accuracy: 0.9983 - precision: 0.9969 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9869 - specificity: 0.9989 - val_loss: 0.0189 - val_tp: 126.0000 - val_fp: 260.0000 - val_tn: 85035.0000 - val_fn: 22.0000 - val_accuracy: 0.9967 - val_precision: 0.3264 - val_recall: 0.8514 - val_auc: 0.9307 - val_sensitivity: 0.8446 - val_specificity: 0.9974
Epoch 20/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0131 - tp: 198956.0000 - fp: 579.0000 - tn: 198441.0000 - fn: 64.0000 - accuracy: 0.9984 - precision: 0.9971 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9878 - specificity: 0.9989 - val_loss: 0.0147 - val_tp: 125.0000 - val_fp: 198.0000 - val_tn: 85097.0000 - val_fn: 23.0000 - val_accuracy: 0.9974 - val_precision: 0.3870 - val_recall: 0.8446 - val_auc: 0.9277 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 21/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0127 - tp: 198965.0000 - fp: 551.0000 - tn: 198469.0000 - fn: 55.0000 - accuracy: 0.9985 - precision: 0.9972 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9881 - specificity: 0.9990 - val_loss: 0.0161 - val_tp: 125.0000 - val_fp: 217.0000 - val_tn: 85078.0000 - val_fn: 23.0000 - val_accuracy: 0.9972 - val_precision: 0.3655 - val_recall: 0.8446 - val_auc: 0.9276 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 22/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0125 - tp: 198966.0000 - fp: 526.0000 - tn: 198494.0000 - fn: 54.0000 - accuracy: 0.9985 - precision: 0.9974 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9879 - specificity: 0.9989 - val_loss: 0.0144 - val_tp: 125.0000 - val_fp: 170.0000 - val_tn: 85125.0000 - val_fn: 23.0000 - val_accuracy: 0.9977 - val_precision: 0.4237 - val_recall: 0.8446 - val_auc: 0.9246 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 23/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0124 - tp: 198956.0000 - fp: 529.0000 - tn: 198491.0000 - fn: 64.0000 - accuracy: 0.9985 - precision: 0.9973 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9886 - specificity: 0.9990 - val_loss: 0.0155 - val_tp: 126.0000 - val_fp: 195.0000 - val_tn: 85100.0000 - val_fn: 22.0000 - val_accuracy: 0.9975 - val_precision: 0.3925 - val_recall: 0.8514 - val_auc: 0.9279 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 24/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0121 - tp: 198957.0000 - fp: 503.0000 - tn: 198517.0000 - fn: 63.0000 - accuracy: 0.9986 - precision: 0.9975 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9896 - specificity: 0.9990 - val_loss: 0.0152 - val_tp: 126.0000 - val_fp: 198.0000 - val_tn: 85097.0000 - val_fn: 22.0000 - val_accuracy: 0.9974 - val_precision: 0.3889 - val_recall: 0.8514 - val_auc: 0.9245 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 25/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0123 - tp: 198966.0000 - fp: 506.0000 - tn: 198514.0000 - fn: 54.0000 - accuracy: 0.9986 - precision: 0.9975 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9885 - specificity: 0.9990 - val_loss: 0.0151 - val_tp: 126.0000 - val_fp: 187.0000 - val_tn: 85108.0000 - val_fn: 22.0000 - val_accuracy: 0.9976 - val_precision: 0.4026 - val_recall: 0.8514 - val_auc: 0.9246 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 26/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0118 - tp: 198968.0000 - fp: 465.0000 - tn: 198555.0000 - fn: 52.0000 - accuracy: 0.9987 - precision: 0.9977 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9896 - specificity: 0.9990 - val_loss: 0.0137 - val_tp: 126.0000 - val_fp: 174.0000 - val_tn: 85121.0000 - val_fn: 22.0000 - val_accuracy: 0.9977 - val_precision: 0.4200 - val_recall: 0.8514 - val_auc: 0.9280 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 27/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0116 - tp: 198963.0000 - fp: 446.0000 - tn: 198574.0000 - fn: 57.0000 - accuracy: 0.9987 - precision: 0.9978 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9898 - specificity: 0.9991 - val_loss: 0.0149 - val_tp: 126.0000 - val_fp: 184.0000 - val_tn: 85111.0000 - val_fn: 22.0000 - val_accuracy: 0.9976 - val_precision: 0.4065 - val_recall: 0.8514 - val_auc: 0.9279 - val_sensitivity: 0.8446 - val_specificity: 0.9988
Epoch 28/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0116 - tp: 198965.0000 - fp: 447.0000 - tn: 198573.0000 - fn: 55.0000 - accuracy: 0.9987 - precision: 0.9978 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9948 - specificity: 0.9991 - val_loss: 0.0134 - val_tp: 126.0000 - val_fp: 152.0000 - val_tn: 85143.0000 - val_fn: 22.0000 - val_accuracy: 0.9980 - val_precision: 0.4532 - val_recall: 0.8514 - val_auc: 0.9280 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 29/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0114 - tp: 198968.0000 - fp: 437.0000 - tn: 198583.0000 - fn: 52.0000 - accuracy: 0.9988 - precision: 0.9978 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9950 - specificity: 0.9991 - val_loss: 0.0134 - val_tp: 126.0000 - val_fp: 144.0000 - val_tn: 85151.0000 - val_fn: 22.0000 - val_accuracy: 0.9981 - val_precision: 0.4667 - val_recall: 0.8514 - val_auc: 0.9247 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 30/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0112 - tp: 198968.0000 - fp: 408.0000 - tn: 198612.0000 - fn: 52.0000 - accuracy: 0.9988 - precision: 0.9980 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9955 - specificity: 0.9991 - val_loss: 0.0125 - val_tp: 126.0000 - val_fp: 135.0000 - val_tn: 85160.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4828 - val_recall: 0.8514 - val_auc: 0.9247 - val_sensitivity: 0.8378 - val_specificity: 0.9989
Epoch 31/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0109 - tp: 198974.0000 - fp: 389.0000 - tn: 198631.0000 - fn: 46.0000 - accuracy: 0.9989 - precision: 0.9980 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9953 - specificity: 0.9991 - val_loss: 0.0116 - val_tp: 126.0000 - val_fp: 116.0000 - val_tn: 85179.0000 - val_fn: 22.0000 - val_accuracy: 0.9984 - val_precision: 0.5207 - val_recall: 0.8514 - val_auc: 0.9248 - val_sensitivity: 0.8378 - val_specificity: 0.9989
Epoch 32/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0116 - tp: 198958.0000 - fp: 445.0000 - tn: 198575.0000 - fn: 62.0000 - accuracy: 0.9987 - precision: 0.9978 - recall: 0.9997 - auc: 0.9994 - sensitivity: 0.9909 - specificity: 0.9990 - val_loss: 0.0121 - val_tp: 125.0000 - val_fp: 128.0000 - val_tn: 85167.0000 - val_fn: 23.0000 - val_accuracy: 0.9982 - val_precision: 0.4941 - val_recall: 0.8446 - val_auc: 0.9247 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 33/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0109 - tp: 198967.0000 - fp: 399.0000 - tn: 198621.0000 - fn: 53.0000 - accuracy: 0.9989 - precision: 0.9980 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9955 - specificity: 0.9991 - val_loss: 0.0136 - val_tp: 125.0000 - val_fp: 149.0000 - val_tn: 85146.0000 - val_fn: 23.0000 - val_accuracy: 0.9980 - val_precision: 0.4562 - val_recall: 0.8446 - val_auc: 0.9247 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 34/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0105 - tp: 198968.0000 - fp: 398.0000 - tn: 198622.0000 - fn: 52.0000 - accuracy: 0.9989 - precision: 0.9980 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9977 - specificity: 0.9992 - val_loss: 0.0115 - val_tp: 125.0000 - val_fp: 114.0000 - val_tn: 85181.0000 - val_fn: 23.0000 - val_accuracy: 0.9984 - val_precision: 0.5230 - val_recall: 0.8446 - val_auc: 0.9249 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 35/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0110 - tp: 198959.0000 - fp: 402.0000 - tn: 198618.0000 - fn: 61.0000 - accuracy: 0.9988 - precision: 0.9980 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9956 - specificity: 0.9991 - val_loss: 0.0148 - val_tp: 126.0000 - val_fp: 182.0000 - val_tn: 85113.0000 - val_fn: 22.0000 - val_accuracy: 0.9976 - val_precision: 0.4091 - val_recall: 0.8514 - val_auc: 0.9279 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 36/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0107 - tp: 198961.0000 - fp: 402.0000 - tn: 198618.0000 - fn: 59.0000 - accuracy: 0.9988 - precision: 0.9980 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9958 - specificity: 0.9991 - val_loss: 0.0125 - val_tp: 126.0000 - val_fp: 131.0000 - val_tn: 85164.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4903 - val_recall: 0.8514 - val_auc: 0.9280 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 37/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0102 - tp: 198975.0000 - fp: 353.0000 - tn: 198667.0000 - fn: 45.0000 - accuracy: 0.9990 - precision: 0.9982 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9971 - specificity: 0.9991 - val_loss: 0.0124 - val_tp: 126.0000 - val_fp: 125.0000 - val_tn: 85170.0000 - val_fn: 22.0000 - val_accuracy: 0.9983 - val_precision: 0.5020 - val_recall: 0.8514 - val_auc: 0.9281 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 38/200
398040/398040 [==============================] - 4s 9us/sample - loss: 0.0104 - tp: 198965.0000 - fp: 363.0000 - tn: 198657.0000 - fn: 55.0000 - accuracy: 0.9989 - precision: 0.9982 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9973 - specificity: 0.9991 - val_loss: 0.0145 - val_tp: 126.0000 - val_fp: 173.0000 - val_tn: 85122.0000 - val_fn: 22.0000 - val_accuracy: 0.9977 - val_precision: 0.4214 - val_recall: 0.8514 - val_auc: 0.9279 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 39/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0104 - tp: 198966.0000 - fp: 388.0000 - tn: 198632.0000 - fn: 54.0000 - accuracy: 0.9989 - precision: 0.9981 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9978 - specificity: 0.9992 - val_loss: 0.0136 - val_tp: 126.0000 - val_fp: 148.0000 - val_tn: 85147.0000 - val_fn: 22.0000 - val_accuracy: 0.9980 - val_precision: 0.4599 - val_recall: 0.8514 - val_auc: 0.9281 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 40/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0099 - tp: 198976.0000 - fp: 340.0000 - tn: 198680.0000 - fn: 44.0000 - accuracy: 0.9990 - precision: 0.9983 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9983 - specificity: 0.9992 - val_loss: 0.0124 - val_tp: 126.0000 - val_fp: 131.0000 - val_tn: 85164.0000 - val_fn: 22.0000 - val_accuracy: 0.9982 - val_precision: 0.4903 - val_recall: 0.8514 - val_auc: 0.9281 - val_sensitivity: 0.8378 - val_specificity: 0.9988
Epoch 41/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0101 - tp: 198970.0000 - fp: 332.0000 - tn: 198688.0000 - fn: 50.0000 - accuracy: 0.9990 - precision: 0.9983 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9987 - specificity: 0.9992 - val_loss: 0.0135 - val_tp: 126.0000 - val_fp: 151.0000 - val_tn: 85144.0000 - val_fn: 22.0000 - val_accuracy: 0.9980 - val_precision: 0.4549 - val_recall: 0.8514 - val_auc: 0.9280 - val_sensitivity: 0.8378 - val_specificity: 0.9985
Epoch 42/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0096 - tp: 198975.0000 - fp: 338.0000 - tn: 198682.0000 - fn: 45.0000 - accuracy: 0.9990 - precision: 0.9983 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9984 - specificity: 0.9993 - val_loss: 0.0116 - val_tp: 124.0000 - val_fp: 119.0000 - val_tn: 85176.0000 - val_fn: 24.0000 - val_accuracy: 0.9983 - val_precision: 0.5103 - val_recall: 0.8378 - val_auc: 0.9282 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 43/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0100 - tp: 198967.0000 - fp: 360.0000 - tn: 198660.0000 - fn: 53.0000 - accuracy: 0.9990 - precision: 0.9982 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9974 - specificity: 0.9991 - val_loss: 0.0108 - val_tp: 124.0000 - val_fp: 104.0000 - val_tn: 85191.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5439 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 44/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0101 - tp: 198970.0000 - fp: 353.0000 - tn: 198667.0000 - fn: 50.0000 - accuracy: 0.9990 - precision: 0.9982 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9973 - specificity: 0.9992 - val_loss: 0.0122 - val_tp: 125.0000 - val_fp: 127.0000 - val_tn: 85168.0000 - val_fn: 23.0000 - val_accuracy: 0.9982 - val_precision: 0.4960 - val_recall: 0.8446 - val_auc: 0.9281 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 45/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0095 - tp: 198975.0000 - fp: 335.0000 - tn: 198685.0000 - fn: 45.0000 - accuracy: 0.9990 - precision: 0.9983 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9985 - specificity: 0.9992 - val_loss: 0.0120 - val_tp: 125.0000 - val_fp: 114.0000 - val_tn: 85181.0000 - val_fn: 23.0000 - val_accuracy: 0.9984 - val_precision: 0.5230 - val_recall: 0.8446 - val_auc: 0.9249 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 46/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0096 - tp: 198975.0000 - fp: 316.0000 - tn: 198704.0000 - fn: 45.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9990 - specificity: 0.9992 - val_loss: 0.0108 - val_tp: 124.0000 - val_fp: 104.0000 - val_tn: 85191.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5439 - val_recall: 0.8378 - val_auc: 0.9249 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 47/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0094 - tp: 198977.0000 - fp: 334.0000 - tn: 198686.0000 - fn: 43.0000 - accuracy: 0.9991 - precision: 0.9983 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9990 - specificity: 0.9993 - val_loss: 0.0117 - val_tp: 125.0000 - val_fp: 113.0000 - val_tn: 85182.0000 - val_fn: 23.0000 - val_accuracy: 0.9984 - val_precision: 0.5252 - val_recall: 0.8446 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 48/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0094 - tp: 198969.0000 - fp: 333.0000 - tn: 198687.0000 - fn: 51.0000 - accuracy: 0.9990 - precision: 0.9983 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9989 - specificity: 0.9992 - val_loss: 0.0116 - val_tp: 125.0000 - val_fp: 118.0000 - val_tn: 85177.0000 - val_fn: 23.0000 - val_accuracy: 0.9983 - val_precision: 0.5144 - val_recall: 0.8446 - val_auc: 0.9282 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 49/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0096 - tp: 198972.0000 - fp: 326.0000 - tn: 198694.0000 - fn: 48.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9988 - specificity: 0.9992 - val_loss: 0.0122 - val_tp: 124.0000 - val_fp: 120.0000 - val_tn: 85175.0000 - val_fn: 24.0000 - val_accuracy: 0.9983 - val_precision: 0.5082 - val_recall: 0.8378 - val_auc: 0.9249 - val_sensitivity: 0.8378 - val_specificity: 0.9985
Epoch 50/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0097 - tp: 198968.0000 - fp: 310.0000 - tn: 198710.0000 - fn: 52.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9991 - specificity: 0.9992 - val_loss: 0.0109 - val_tp: 124.0000 - val_fp: 107.0000 - val_tn: 85188.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5368 - val_recall: 0.8378 - val_auc: 0.9282 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 51/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0094 - tp: 198966.0000 - fp: 314.0000 - tn: 198706.0000 - fn: 54.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9991 - specificity: 0.9993 - val_loss: 0.0113 - val_tp: 124.0000 - val_fp: 109.0000 - val_tn: 85186.0000 - val_fn: 24.0000 - val_accuracy: 0.9984 - val_precision: 0.5322 - val_recall: 0.8378 - val_auc: 0.9282 - val_sensitivity: 0.8378 - val_specificity: 0.9984
Epoch 52/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0094 - tp: 198967.0000 - fp: 319.0000 - tn: 198701.0000 - fn: 53.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9992 - specificity: 0.9992 - val_loss: 0.0111 - val_tp: 124.0000 - val_fp: 105.0000 - val_tn: 85190.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5415 - val_recall: 0.8378 - val_auc: 0.9249 - val_sensitivity: 0.8378 - val_specificity: 0.9985
Epoch 53/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0095 - tp: 198969.0000 - fp: 302.0000 - tn: 198718.0000 - fn: 51.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9989 - specificity: 0.9993 - val_loss: 0.0112 - val_tp: 124.0000 - val_fp: 103.0000 - val_tn: 85192.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5463 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 54/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0099 - tp: 198968.0000 - fp: 334.0000 - tn: 198686.0000 - fn: 52.0000 - accuracy: 0.9990 - precision: 0.9983 - recall: 0.9997 - auc: 0.9995 - sensitivity: 0.9989 - specificity: 0.9992 - val_loss: 0.0108 - val_tp: 124.0000 - val_fp: 102.0000 - val_tn: 85193.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5487 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 55/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0091 - tp: 198970.0000 - fp: 306.0000 - tn: 198714.0000 - fn: 50.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9994 - specificity: 0.9993 - val_loss: 0.0109 - val_tp: 124.0000 - val_fp: 104.0000 - val_tn: 85191.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5439 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 56/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0094 - tp: 198975.0000 - fp: 317.0000 - tn: 198703.0000 - fn: 45.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9998 - auc: 0.9995 - sensitivity: 0.9990 - specificity: 0.9993 - val_loss: 0.0111 - val_tp: 124.0000 - val_fp: 106.0000 - val_tn: 85189.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5391 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9987
Epoch 57/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0094 - tp: 198978.0000 - fp: 296.0000 - tn: 198724.0000 - fn: 42.0000 - accuracy: 0.9992 - precision: 0.9985 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9992 - specificity: 0.9993 - val_loss: 0.0105 - val_tp: 124.0000 - val_fp: 96.0000 - val_tn: 85199.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5636 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9984
Epoch 58/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0091 - tp: 198974.0000 - fp: 319.0000 - tn: 198701.0000 - fn: 46.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9994 - specificity: 0.9994 - val_loss: 0.0111 - val_tp: 124.0000 - val_fp: 103.0000 - val_tn: 85192.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5463 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 59/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0095 - tp: 198970.0000 - fp: 290.0000 - tn: 198730.0000 - fn: 50.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9993 - specificity: 0.9993 - val_loss: 0.0106 - val_tp: 124.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5586 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9984
Epoch 60/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0090 - tp: 198973.0000 - fp: 286.0000 - tn: 198734.0000 - fn: 47.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0115 - val_tp: 124.0000 - val_fp: 106.0000 - val_tn: 85189.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5391 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9984
Epoch 61/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0092 - tp: 198969.0000 - fp: 322.0000 - tn: 198698.0000 - fn: 51.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9990 - specificity: 0.9993 - val_loss: 0.0107 - val_tp: 124.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5586 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9985
Epoch 62/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0095 - tp: 198967.0000 - fp: 305.0000 - tn: 198715.0000 - fn: 53.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9992 - specificity: 0.9993 - val_loss: 0.0102 - val_tp: 124.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5586 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9986
Epoch 63/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0092 - tp: 198973.0000 - fp: 307.0000 - tn: 198713.0000 - fn: 47.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9992 - specificity: 0.9993 - val_loss: 0.0109 - val_tp: 124.0000 - val_fp: 101.0000 - val_tn: 85194.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5511 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 64/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0090 - tp: 198976.0000 - fp: 311.0000 - tn: 198709.0000 - fn: 44.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0097 - val_tp: 124.0000 - val_fp: 91.0000 - val_tn: 85204.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5767 - val_recall: 0.8378 - val_auc: 0.9251 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 65/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0089 - tp: 198972.0000 - fp: 284.0000 - tn: 198736.0000 - fn: 48.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0106 - val_tp: 124.0000 - val_fp: 102.0000 - val_tn: 85193.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5487 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 66/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0090 - tp: 198975.0000 - fp: 285.0000 - tn: 198735.0000 - fn: 45.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9994 - val_loss: 0.0103 - val_tp: 124.0000 - val_fp: 96.0000 - val_tn: 85199.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5636 - val_recall: 0.8378 - val_auc: 0.9250 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 67/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0091 - tp: 198973.0000 - fp: 278.0000 - tn: 198742.0000 - fn: 47.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0101 - val_tp: 124.0000 - val_fp: 95.0000 - val_tn: 85200.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5662 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 68/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0090 - tp: 198974.0000 - fp: 293.0000 - tn: 198727.0000 - fn: 46.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9994 - specificity: 0.9993 - val_loss: 0.0102 - val_tp: 124.0000 - val_fp: 96.0000 - val_tn: 85199.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5636 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 69/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0086 - tp: 198972.0000 - fp: 253.0000 - tn: 198767.0000 - fn: 48.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0111 - val_tp: 124.0000 - val_fp: 105.0000 - val_tn: 85190.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5415 - val_recall: 0.8378 - val_auc: 0.9316 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 70/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0087 - tp: 198974.0000 - fp: 294.0000 - tn: 198726.0000 - fn: 46.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9994 - specificity: 0.9993 - val_loss: 0.0099 - val_tp: 124.0000 - val_fp: 94.0000 - val_tn: 85201.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5688 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 71/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0088 - tp: 198971.0000 - fp: 263.0000 - tn: 198757.0000 - fn: 49.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9993 - val_loss: 0.0102 - val_tp: 124.0000 - val_fp: 93.0000 - val_tn: 85202.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5714 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 72/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0088 - tp: 198973.0000 - fp: 297.0000 - tn: 198723.0000 - fn: 47.0000 - accuracy: 0.9991 - precision: 0.9985 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0105 - val_tp: 124.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5586 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9992
Epoch 73/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0089 - tp: 198971.0000 - fp: 314.0000 - tn: 198706.0000 - fn: 49.0000 - accuracy: 0.9991 - precision: 0.9984 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0097 - val_tp: 124.0000 - val_fp: 89.0000 - val_tn: 85206.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5822 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 74/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0087 - tp: 198971.0000 - fp: 273.0000 - tn: 198747.0000 - fn: 49.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0109 - val_tp: 123.0000 - val_fp: 97.0000 - val_tn: 85198.0000 - val_fn: 25.0000 - val_accuracy: 0.9986 - val_precision: 0.5591 - val_recall: 0.8311 - val_auc: 0.9283 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 75/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0087 - tp: 198978.0000 - fp: 264.0000 - tn: 198756.0000 - fn: 42.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9993 - val_loss: 0.0098 - val_tp: 124.0000 - val_fp: 85.0000 - val_tn: 85210.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5933 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 76/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0083 - tp: 198973.0000 - fp: 265.0000 - tn: 198755.0000 - fn: 47.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0106 - val_tp: 124.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5586 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 77/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0085 - tp: 198974.0000 - fp: 258.0000 - tn: 198762.0000 - fn: 46.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0103 - val_tp: 124.0000 - val_fp: 95.0000 - val_tn: 85200.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5662 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 78/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0085 - tp: 198972.0000 - fp: 273.0000 - tn: 198747.0000 - fn: 48.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0098 - val_tp: 124.0000 - val_fp: 87.0000 - val_tn: 85208.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5877 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 79/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0087 - tp: 198974.0000 - fp: 269.0000 - tn: 198751.0000 - fn: 46.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9993 - val_loss: 0.0106 - val_tp: 124.0000 - val_fp: 102.0000 - val_tn: 85193.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5487 - val_recall: 0.8378 - val_auc: 0.9283 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 80/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0087 - tp: 198971.0000 - fp: 265.0000 - tn: 198755.0000 - fn: 49.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9993 - val_loss: 0.0096 - val_tp: 124.0000 - val_fp: 85.0000 - val_tn: 85210.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5933 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 81/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0085 - tp: 198974.0000 - fp: 254.0000 - tn: 198766.0000 - fn: 46.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0099 - val_tp: 124.0000 - val_fp: 91.0000 - val_tn: 85204.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5767 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 82/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0088 - tp: 198973.0000 - fp: 267.0000 - tn: 198753.0000 - fn: 47.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0095 - val_tp: 124.0000 - val_fp: 84.0000 - val_tn: 85211.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5962 - val_recall: 0.8378 - val_auc: 0.9285 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 83/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0083 - tp: 198975.0000 - fp: 239.0000 - tn: 198781.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0107 - val_tp: 124.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5586 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 84/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0087 - tp: 198977.0000 - fp: 250.0000 - tn: 198770.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9994 - val_loss: 0.0098 - val_tp: 123.0000 - val_fp: 85.0000 - val_tn: 85210.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5913 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 85/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0086 - tp: 198970.0000 - fp: 268.0000 - tn: 198752.0000 - fn: 50.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9993 - val_loss: 0.0100 - val_tp: 123.0000 - val_fp: 92.0000 - val_tn: 85203.0000 - val_fn: 25.0000 - val_accuracy: 0.9986 - val_precision: 0.5721 - val_recall: 0.8311 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 86/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0088 - tp: 198974.0000 - fp: 272.0000 - tn: 198748.0000 - fn: 46.0000 - accuracy: 0.9992 - precision: 0.9986 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9994 - val_loss: 0.0100 - val_tp: 123.0000 - val_fp: 86.0000 - val_tn: 85209.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5885 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 87/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0086 - tp: 198974.0000 - fp: 249.0000 - tn: 198771.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0095 - val_tp: 123.0000 - val_fp: 85.0000 - val_tn: 85210.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5913 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 88/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0083 - tp: 198977.0000 - fp: 241.0000 - tn: 198779.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0094 - val_tp: 123.0000 - val_fp: 82.0000 - val_tn: 85213.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.6000 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 89/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0089 - tp: 198976.0000 - fp: 263.0000 - tn: 198757.0000 - fn: 44.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9995 - specificity: 0.9993 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 81.0000 - val_tn: 85214.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6029 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 90/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0083 - tp: 198976.0000 - fp: 242.0000 - tn: 198778.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0100 - val_tp: 124.0000 - val_fp: 88.0000 - val_tn: 85207.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5849 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 91/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0083 - tp: 198973.0000 - fp: 248.0000 - tn: 198772.0000 - fn: 47.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0106 - val_tp: 124.0000 - val_fp: 99.0000 - val_tn: 85196.0000 - val_fn: 24.0000 - val_accuracy: 0.9986 - val_precision: 0.5561 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 92/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198975.0000 - fp: 229.0000 - tn: 198791.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0110 - val_tp: 124.0000 - val_fp: 100.0000 - val_tn: 85195.0000 - val_fn: 24.0000 - val_accuracy: 0.9985 - val_precision: 0.5536 - val_recall: 0.8378 - val_auc: 0.9317 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 93/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0081 - tp: 198975.0000 - fp: 239.0000 - tn: 198781.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0098 - val_tp: 123.0000 - val_fp: 90.0000 - val_tn: 85205.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5775 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 94/200
398040/398040 [==============================] - 5s 14us/sample - loss: 0.0085 - tp: 198970.0000 - fp: 252.0000 - tn: 198768.0000 - fn: 50.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0099 - val_tp: 123.0000 - val_fp: 91.0000 - val_tn: 85204.0000 - val_fn: 25.0000 - val_accuracy: 0.9986 - val_precision: 0.5748 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 95/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0078 - tp: 198974.0000 - fp: 244.0000 - tn: 198776.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0098 - val_tp: 123.0000 - val_fp: 82.0000 - val_tn: 85213.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.6000 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 96/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0083 - tp: 198971.0000 - fp: 251.0000 - tn: 198769.0000 - fn: 49.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0096 - val_tp: 123.0000 - val_fp: 83.0000 - val_tn: 85212.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5971 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 97/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0083 - tp: 198978.0000 - fp: 241.0000 - tn: 198779.0000 - fn: 42.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 98/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0081 - tp: 198975.0000 - fp: 241.0000 - tn: 198779.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0106 - val_tp: 123.0000 - val_fp: 98.0000 - val_tn: 85197.0000 - val_fn: 25.0000 - val_accuracy: 0.9986 - val_precision: 0.5566 - val_recall: 0.8311 - val_auc: 0.9283 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 99/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0079 - tp: 198976.0000 - fp: 253.0000 - tn: 198767.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0103 - val_tp: 123.0000 - val_fp: 88.0000 - val_tn: 85207.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5829 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 100/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0078 - tp: 198976.0000 - fp: 223.0000 - tn: 198797.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0100 - val_tp: 123.0000 - val_fp: 89.0000 - val_tn: 85206.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5802 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 101/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0085 - tp: 198970.0000 - fp: 257.0000 - tn: 198763.0000 - fn: 50.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0097 - val_tp: 123.0000 - val_fp: 83.0000 - val_tn: 85212.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5971 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 102/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0081 - tp: 198976.0000 - fp: 235.0000 - tn: 198785.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0102 - val_tp: 123.0000 - val_fp: 94.0000 - val_tn: 85201.0000 - val_fn: 25.0000 - val_accuracy: 0.9986 - val_precision: 0.5668 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9986
Epoch 103/200
398040/398040 [==============================] - 5s 13us/sample - loss: 0.0082 - tp: 198975.0000 - fp: 249.0000 - tn: 198771.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0099 - val_tp: 123.0000 - val_fp: 88.0000 - val_tn: 85207.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5829 - val_recall: 0.8311 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 104/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0080 - tp: 198975.0000 - fp: 219.0000 - tn: 198801.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0094 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 105/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0082 - tp: 198972.0000 - fp: 232.0000 - tn: 198788.0000 - fn: 48.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 106/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0084 - tp: 198971.0000 - fp: 258.0000 - tn: 198762.0000 - fn: 49.0000 - accuracy: 0.9992 - precision: 0.9987 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0092 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 107/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198974.0000 - fp: 223.0000 - tn: 198797.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0095 - val_tp: 123.0000 - val_fp: 81.0000 - val_tn: 85214.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6029 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 108/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198977.0000 - fp: 223.0000 - tn: 198797.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0088 - val_tp: 123.0000 - val_fp: 71.0000 - val_tn: 85224.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6340 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 109/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0079 - tp: 198976.0000 - fp: 241.0000 - tn: 198779.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 110/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198976.0000 - fp: 228.0000 - tn: 198792.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 75.0000 - val_tn: 85220.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6212 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 111/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198972.0000 - fp: 221.0000 - tn: 198799.0000 - fn: 48.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0088 - val_tp: 123.0000 - val_fp: 71.0000 - val_tn: 85224.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6340 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 112/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198977.0000 - fp: 244.0000 - tn: 198776.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9996 - specificity: 0.9994 - val_loss: 0.0098 - val_tp: 123.0000 - val_fp: 81.0000 - val_tn: 85214.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6029 - val_recall: 0.8311 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 113/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198980.0000 - fp: 208.0000 - tn: 198812.0000 - fn: 40.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0101 - val_tp: 123.0000 - val_fp: 95.0000 - val_tn: 85200.0000 - val_fn: 25.0000 - val_accuracy: 0.9986 - val_precision: 0.5642 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 114/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198977.0000 - fp: 212.0000 - tn: 198808.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0102 - val_tp: 123.0000 - val_fp: 89.0000 - val_tn: 85206.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5802 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 115/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0082 - tp: 198975.0000 - fp: 236.0000 - tn: 198784.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0084 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 116/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198980.0000 - fp: 211.0000 - tn: 198809.0000 - fn: 40.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 117/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198968.0000 - fp: 211.0000 - tn: 198809.0000 - fn: 52.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0102 - val_tp: 124.0000 - val_fp: 86.0000 - val_tn: 85209.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5905 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 118/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198977.0000 - fp: 228.0000 - tn: 198792.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0094 - val_tp: 123.0000 - val_fp: 79.0000 - val_tn: 85216.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6089 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 119/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0082 - tp: 198979.0000 - fp: 230.0000 - tn: 198790.0000 - fn: 41.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 120/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198975.0000 - fp: 216.0000 - tn: 198804.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0100 - val_tp: 123.0000 - val_fp: 88.0000 - val_tn: 85207.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5829 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 121/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198974.0000 - fp: 222.0000 - tn: 198798.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0094 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 122/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198974.0000 - fp: 225.0000 - tn: 198795.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 123/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0079 - tp: 198973.0000 - fp: 230.0000 - tn: 198790.0000 - fn: 47.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 79.0000 - val_tn: 85216.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6089 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 124/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198977.0000 - fp: 226.0000 - tn: 198794.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 125/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198975.0000 - fp: 210.0000 - tn: 198810.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 69.0000 - val_tn: 85226.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6406 - val_recall: 0.8311 - val_auc: 0.9286 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 126/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198970.0000 - fp: 231.0000 - tn: 198789.0000 - fn: 50.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9997 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0098 - val_tp: 124.0000 - val_fp: 89.0000 - val_tn: 85206.0000 - val_fn: 24.0000 - val_accuracy: 0.9987 - val_precision: 0.5822 - val_recall: 0.8378 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 127/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198977.0000 - fp: 217.0000 - tn: 198803.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 82.0000 - val_tn: 85213.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.6000 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 128/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198975.0000 - fp: 218.0000 - tn: 198802.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0097 - val_tp: 123.0000 - val_fp: 82.0000 - val_tn: 85213.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.6000 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 129/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198971.0000 - fp: 230.0000 - tn: 198790.0000 - fn: 49.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 80.0000 - val_tn: 85215.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6059 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 130/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198976.0000 - fp: 228.0000 - tn: 198792.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 77.0000 - val_tn: 85218.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6150 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 131/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198974.0000 - fp: 214.0000 - tn: 198806.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0088 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 132/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0076 - tp: 198977.0000 - fp: 218.0000 - tn: 198802.0000 - fn: 43.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 80.0000 - val_tn: 85215.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6059 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9985
Epoch 133/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198978.0000 - fp: 217.0000 - tn: 198803.0000 - fn: 42.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0090 - val_tp: 123.0000 - val_fp: 77.0000 - val_tn: 85218.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6150 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 134/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198974.0000 - fp: 221.0000 - tn: 198799.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0097 - val_tp: 123.0000 - val_fp: 86.0000 - val_tn: 85209.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5885 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 135/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0081 - tp: 198973.0000 - fp: 223.0000 - tn: 198797.0000 - fn: 47.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 66.0000 - val_tn: 85229.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6508 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 136/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0076 - tp: 198977.0000 - fp: 211.0000 - tn: 198809.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 137/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198976.0000 - fp: 220.0000 - tn: 198800.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 138/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198976.0000 - fp: 231.0000 - tn: 198789.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0096 - val_tp: 123.0000 - val_fp: 81.0000 - val_tn: 85214.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6029 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 139/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0078 - tp: 198975.0000 - fp: 223.0000 - tn: 198797.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 140/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198974.0000 - fp: 222.0000 - tn: 198798.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 141/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198972.0000 - fp: 213.0000 - tn: 198807.0000 - fn: 48.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9994 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 69.0000 - val_tn: 85226.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6406 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 142/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0074 - tp: 198980.0000 - fp: 194.0000 - tn: 198826.0000 - fn: 40.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 143/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198976.0000 - fp: 220.0000 - tn: 198800.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 144/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198975.0000 - fp: 213.0000 - tn: 198807.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 69.0000 - val_tn: 85226.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6406 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 145/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198977.0000 - fp: 201.0000 - tn: 198819.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0091 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 146/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0078 - tp: 198974.0000 - fp: 219.0000 - tn: 198801.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0083 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 147/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198979.0000 - fp: 220.0000 - tn: 198800.0000 - fn: 41.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 76.0000 - val_tn: 85219.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6181 - val_recall: 0.8311 - val_auc: 0.9319 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 148/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198978.0000 - fp: 211.0000 - tn: 198809.0000 - fn: 42.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 149/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198979.0000 - fp: 229.0000 - tn: 198791.0000 - fn: 41.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 150/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198979.0000 - fp: 194.0000 - tn: 198826.0000 - fn: 41.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 69.0000 - val_tn: 85226.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6406 - val_recall: 0.8311 - val_auc: 0.9185 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 151/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0079 - tp: 198971.0000 - fp: 232.0000 - tn: 198788.0000 - fn: 49.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0091 - val_tp: 123.0000 - val_fp: 80.0000 - val_tn: 85215.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6059 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 152/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0076 - tp: 198974.0000 - fp: 221.0000 - tn: 198799.0000 - fn: 46.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9986
Epoch 153/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198973.0000 - fp: 238.0000 - tn: 198782.0000 - fn: 47.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9996 - sensitivity: 0.9997 - specificity: 0.9994 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 154/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0074 - tp: 198980.0000 - fp: 189.0000 - tn: 198831.0000 - fn: 40.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 155/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0073 - tp: 198975.0000 - fp: 197.0000 - tn: 198823.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 72.0000 - val_tn: 85223.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6308 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 156/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198971.0000 - fp: 202.0000 - tn: 198818.0000 - fn: 49.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 157/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0080 - tp: 198973.0000 - fp: 234.0000 - tn: 198786.0000 - fn: 47.0000 - accuracy: 0.9993 - precision: 0.9988 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0094 - val_tp: 123.0000 - val_fp: 84.0000 - val_tn: 85211.0000 - val_fn: 25.0000 - val_accuracy: 0.9987 - val_precision: 0.5942 - val_recall: 0.8311 - val_auc: 0.9284 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 158/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198981.0000 - fp: 206.0000 - tn: 198814.0000 - fn: 39.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0080 - val_tp: 123.0000 - val_fp: 60.0000 - val_tn: 85235.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6721 - val_recall: 0.8311 - val_auc: 0.9185 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 159/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0073 - tp: 198978.0000 - fp: 203.0000 - tn: 198817.0000 - fn: 42.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0090 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 160/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198972.0000 - fp: 209.0000 - tn: 198811.0000 - fn: 48.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0095 - val_tp: 123.0000 - val_fp: 81.0000 - val_tn: 85214.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6029 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 161/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0077 - tp: 198979.0000 - fp: 207.0000 - tn: 198813.0000 - fn: 41.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9286 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 162/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0075 - tp: 198976.0000 - fp: 212.0000 - tn: 198808.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 163/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.0073 - tp: 198974.0000 - fp: 199.0000 - tn: 198821.0000 - fn: 46.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 164/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0075 - tp: 198976.0000 - fp: 207.0000 - tn: 198813.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 65.0000 - val_tn: 85230.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6543 - val_recall: 0.8311 - val_auc: 0.9185 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 165/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198972.0000 - fp: 220.0000 - tn: 198800.0000 - fn: 48.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0088 - val_tp: 123.0000 - val_fp: 69.0000 - val_tn: 85226.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6406 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 166/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198976.0000 - fp: 213.0000 - tn: 198807.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 79.0000 - val_tn: 85216.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6089 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 167/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0074 - tp: 198973.0000 - fp: 204.0000 - tn: 198816.0000 - fn: 47.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0090 - val_tp: 123.0000 - val_fp: 72.0000 - val_tn: 85223.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6308 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 168/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0077 - tp: 198975.0000 - fp: 189.0000 - tn: 198831.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 77.0000 - val_tn: 85218.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6150 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 169/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0073 - tp: 198976.0000 - fp: 213.0000 - tn: 198807.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 66.0000 - val_tn: 85229.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6508 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 170/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0074 - tp: 198977.0000 - fp: 205.0000 - tn: 198815.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 65.0000 - val_tn: 85230.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6543 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 171/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198977.0000 - fp: 202.0000 - tn: 198818.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0090 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 172/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0076 - tp: 198974.0000 - fp: 208.0000 - tn: 198812.0000 - fn: 46.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0085 - val_tp: 123.0000 - val_fp: 67.0000 - val_tn: 85228.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6474 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 173/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0078 - tp: 198978.0000 - fp: 213.0000 - tn: 198807.0000 - fn: 42.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0083 - val_tp: 123.0000 - val_fp: 61.0000 - val_tn: 85234.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6685 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 174/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0074 - tp: 198975.0000 - fp: 208.0000 - tn: 198812.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 175/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0073 - tp: 198976.0000 - fp: 189.0000 - tn: 198831.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0082 - val_tp: 123.0000 - val_fp: 60.0000 - val_tn: 85235.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6721 - val_recall: 0.8311 - val_auc: 0.9185 - val_sensitivity: 0.8311 - val_specificity: 0.9987
Epoch 176/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0074 - tp: 198974.0000 - fp: 204.0000 - tn: 198816.0000 - fn: 46.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0088 - val_tp: 123.0000 - val_fp: 72.0000 - val_tn: 85223.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6308 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 177/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.0072 - tp: 198977.0000 - fp: 194.0000 - tn: 198826.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0093 - val_tp: 123.0000 - val_fp: 75.0000 - val_tn: 85220.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6212 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 178/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0077 - tp: 198976.0000 - fp: 217.0000 - tn: 198803.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0097 - val_tp: 124.0000 - val_fp: 81.0000 - val_tn: 85214.0000 - val_fn: 24.0000 - val_accuracy: 0.9988 - val_precision: 0.6049 - val_recall: 0.8378 - val_auc: 0.9285 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 179/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0078 - tp: 198970.0000 - fp: 216.0000 - tn: 198804.0000 - fn: 50.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0088 - val_tp: 123.0000 - val_fp: 72.0000 - val_tn: 85223.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6308 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 180/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0070 - tp: 198980.0000 - fp: 174.0000 - tn: 198846.0000 - fn: 40.0000 - accuracy: 0.9995 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 65.0000 - val_tn: 85230.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6543 - val_recall: 0.8311 - val_auc: 0.9185 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 181/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198974.0000 - fp: 195.0000 - tn: 198825.0000 - fn: 46.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 70.0000 - val_tn: 85225.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6373 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 182/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198977.0000 - fp: 212.0000 - tn: 198808.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 66.0000 - val_tn: 85229.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6508 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 183/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198980.0000 - fp: 198.0000 - tn: 198822.0000 - fn: 40.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0091 - val_tp: 123.0000 - val_fp: 71.0000 - val_tn: 85224.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6340 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 184/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198976.0000 - fp: 191.0000 - tn: 198829.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 72.0000 - val_tn: 85223.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6308 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 185/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198976.0000 - fp: 223.0000 - tn: 198797.0000 - fn: 44.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0091 - val_tp: 123.0000 - val_fp: 74.0000 - val_tn: 85221.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6244 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 186/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0074 - tp: 198975.0000 - fp: 216.0000 - tn: 198804.0000 - fn: 45.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 78.0000 - val_tn: 85217.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6119 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 187/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198976.0000 - fp: 191.0000 - tn: 198829.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0092 - val_tp: 123.0000 - val_fp: 79.0000 - val_tn: 85216.0000 - val_fn: 25.0000 - val_accuracy: 0.9988 - val_precision: 0.6089 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 188/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0075 - tp: 198974.0000 - fp: 210.0000 - tn: 198810.0000 - fn: 46.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0083 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9319 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 189/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0073 - tp: 198979.0000 - fp: 185.0000 - tn: 198835.0000 - fn: 41.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 68.0000 - val_tn: 85227.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6440 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8378 - val_specificity: 0.9990
Epoch 190/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198976.0000 - fp: 212.0000 - tn: 198808.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0084 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 191/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0077 - tp: 198972.0000 - fp: 210.0000 - tn: 198810.0000 - fn: 48.0000 - accuracy: 0.9994 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0084 - val_tp: 123.0000 - val_fp: 62.0000 - val_tn: 85233.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6649 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 192/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0076 - tp: 198970.0000 - fp: 208.0000 - tn: 198812.0000 - fn: 50.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0089 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 193/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198981.0000 - fp: 193.0000 - tn: 198827.0000 - fn: 39.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 65.0000 - val_tn: 85230.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6543 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 194/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198978.0000 - fp: 186.0000 - tn: 198834.0000 - fn: 42.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0084 - val_tp: 123.0000 - val_fp: 59.0000 - val_tn: 85236.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6758 - val_recall: 0.8311 - val_auc: 0.9218 - val_sensitivity: 0.8378 - val_specificity: 0.9991
Epoch 195/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0074 - tp: 198969.0000 - fp: 221.0000 - tn: 198799.0000 - fn: 51.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9997 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 196/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198977.0000 - fp: 192.0000 - tn: 198828.0000 - fn: 43.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 64.0000 - val_tn: 85231.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6578 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 197/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0072 - tp: 198975.0000 - fp: 186.0000 - tn: 198834.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9991 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0090 - val_tp: 123.0000 - val_fp: 73.0000 - val_tn: 85222.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6276 - val_recall: 0.8311 - val_auc: 0.9318 - val_sensitivity: 0.8311 - val_specificity: 0.9988
Epoch 198/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0076 - tp: 198973.0000 - fp: 220.0000 - tn: 198800.0000 - fn: 47.0000 - accuracy: 0.9993 - precision: 0.9989 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9997 - specificity: 0.9995 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 67.0000 - val_tn: 85228.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6474 - val_recall: 0.8311 - val_auc: 0.9251 - val_sensitivity: 0.8311 - val_specificity: 0.9990
Epoch 199/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0069 - tp: 198975.0000 - fp: 198.0000 - tn: 198822.0000 - fn: 45.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0087 - val_tp: 123.0000 - val_fp: 66.0000 - val_tn: 85229.0000 - val_fn: 25.0000 - val_accuracy: 0.9989 - val_precision: 0.6508 - val_recall: 0.8311 - val_auc: 0.9285 - val_sensitivity: 0.8311 - val_specificity: 0.9989
Epoch 200/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0070 - tp: 198976.0000 - fp: 195.0000 - tn: 198825.0000 - fn: 44.0000 - accuracy: 0.9994 - precision: 0.9990 - recall: 0.9998 - auc: 0.9997 - sensitivity: 0.9998 - specificity: 0.9996 - val_loss: 0.0086 - val_tp: 123.0000 - val_fp: 62.0000 - val_tn: 85233.0000 - val_fn: 25.0000 - val_accuracy: 0.9990 - val_precision: 0.6649 - val_recall: 0.8311 - val_auc: 0.9252 - val_sensitivity: 0.8311 - val_specificity: 0.9989

#cost_sensitive weights to punish to the false negatives
run_500_weight_history = model_500_weight.fit(x=trainX_scaled,y=trainy_scaled,class_weight={0:1,1:500},validation_data=(testX,testy),batch_size=BATCH_SIZE,epochs=EPOCHS)#,callbacks=[es])

Train on 398040 samples, validate on 85443 samples
Epoch 1/200
398040/398040 [==============================] - 5s 14us/sample - loss: 12.3821 - tp: 195687.0000 - fp: 121798.0000 - tn: 77222.0000 - fn: 3333.0000 - accuracy: 0.6856 - precision: 0.6164 - recall: 0.9833 - auc: 0.9135 - sensitivity: 0.0000e+00 - specificity: 0.8833 - val_loss: 1.6347 - val_tp: 148.0000 - val_fp: 43660.0000 - val_tn: 41635.0000 - val_fn: 0.0000e+00 - val_accuracy: 0.4890 - val_precision: 0.0034 - val_recall: 1.0000 - val_auc: 0.9349 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9151
Epoch 2/200
398040/398040 [==============================] - 4s 11us/sample - loss: 1.0265 - tp: 199011.0000 - fp: 81463.0000 - tn: 117557.0000 - fn: 9.0000 - accuracy: 0.7953 - precision: 0.7096 - recall: 1.0000 - auc: 0.9621 - sensitivity: 0.0000e+00 - specificity: 0.9299 - val_loss: 0.8463 - val_tp: 139.0000 - val_fp: 18931.0000 - val_tn: 66364.0000 - val_fn: 9.0000 - val_accuracy: 0.7783 - val_precision: 0.0073 - val_recall: 0.9392 - val_auc: 0.9525 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9162
Epoch 3/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.6359 - tp: 198997.0000 - fp: 45777.0000 - tn: 153243.0000 - fn: 23.0000 - accuracy: 0.8849 - precision: 0.8130 - recall: 0.9999 - auc: 0.9758 - sensitivity: 0.0000e+00 - specificity: 0.9531 - val_loss: 0.4991 - val_tp: 137.0000 - val_fp: 10472.0000 - val_tn: 74823.0000 - val_fn: 11.0000 - val_accuracy: 0.8773 - val_precision: 0.0129 - val_recall: 0.9257 - val_auc: 0.9583 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9659
Epoch 4/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.4930 - tp: 198999.0000 - fp: 29416.0000 - tn: 169604.0000 - fn: 21.0000 - accuracy: 0.9260 - precision: 0.8712 - recall: 0.9999 - auc: 0.9822 - sensitivity: 0.0000e+00 - specificity: 0.9650 - val_loss: 0.3547 - val_tp: 137.0000 - val_fp: 6592.0000 - val_tn: 78703.0000 - val_fn: 11.0000 - val_accuracy: 0.9227 - val_precision: 0.0204 - val_recall: 0.9257 - val_auc: 0.9577 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9735
Epoch 5/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.4094 - tp: 199000.0000 - fp: 22232.0000 - tn: 176788.0000 - fn: 20.0000 - accuracy: 0.9441 - precision: 0.8995 - recall: 0.9999 - auc: 0.9854 - sensitivity: 0.0000e+00 - specificity: 0.9713 - val_loss: 0.3020 - val_tp: 137.0000 - val_fp: 5258.0000 - val_tn: 80037.0000 - val_fn: 11.0000 - val_accuracy: 0.9383 - val_precision: 0.0254 - val_recall: 0.9257 - val_auc: 0.9583 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9731
Epoch 6/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.3791 - tp: 199000.0000 - fp: 18689.0000 - tn: 180331.0000 - fn: 20.0000 - accuracy: 0.9530 - precision: 0.9141 - recall: 0.9999 - auc: 0.9873 - sensitivity: 0.0000e+00 - specificity: 0.9749 - val_loss: 0.2527 - val_tp: 137.0000 - val_fp: 4618.0000 - val_tn: 80677.0000 - val_fn: 11.0000 - val_accuracy: 0.9458 - val_precision: 0.0288 - val_recall: 0.9257 - val_auc: 0.9592 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9783
Epoch 7/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.3252 - tp: 199000.0000 - fp: 16536.0000 - tn: 182484.0000 - fn: 20.0000 - accuracy: 0.9584 - precision: 0.9233 - recall: 0.9999 - auc: 0.9883 - sensitivity: 0.0000e+00 - specificity: 0.9767 - val_loss: 0.2326 - val_tp: 137.0000 - val_fp: 4074.0000 - val_tn: 81221.0000 - val_fn: 11.0000 - val_accuracy: 0.9522 - val_precision: 0.0325 - val_recall: 0.9257 - val_auc: 0.9611 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9822
Epoch 8/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.3464 - tp: 198994.0000 - fp: 14835.0000 - tn: 184185.0000 - fn: 26.0000 - accuracy: 0.9627 - precision: 0.9306 - recall: 0.9999 - auc: 0.9895 - sensitivity: 0.0000e+00 - specificity: 0.9792 - val_loss: 0.2836 - val_tp: 136.0000 - val_fp: 4877.0000 - val_tn: 80418.0000 - val_fn: 12.0000 - val_accuracy: 0.9428 - val_precision: 0.0271 - val_recall: 0.9189 - val_auc: 0.9583 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9777
Epoch 9/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.3176 - tp: 198998.0000 - fp: 15415.0000 - tn: 183605.0000 - fn: 22.0000 - accuracy: 0.9612 - precision: 0.9281 - recall: 0.9999 - auc: 0.9894 - sensitivity: 0.0000e+00 - specificity: 0.9790 - val_loss: 0.2291 - val_tp: 136.0000 - val_fp: 3918.0000 - val_tn: 81377.0000 - val_fn: 12.0000 - val_accuracy: 0.9540 - val_precision: 0.0335 - val_recall: 0.9189 - val_auc: 0.9614 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9825
Epoch 10/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.3096 - tp: 199000.0000 - fp: 13889.0000 - tn: 185131.0000 - fn: 20.0000 - accuracy: 0.9651 - precision: 0.9348 - recall: 0.9999 - auc: 0.9904 - sensitivity: 0.0000e+00 - specificity: 0.9809 - val_loss: 0.2280 - val_tp: 136.0000 - val_fp: 3833.0000 - val_tn: 81462.0000 - val_fn: 12.0000 - val_accuracy: 0.9550 - val_precision: 0.0343 - val_recall: 0.9189 - val_auc: 0.9615 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9746
Epoch 11/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2970 - tp: 198999.0000 - fp: 12718.0000 - tn: 186302.0000 - fn: 21.0000 - accuracy: 0.9680 - precision: 0.9399 - recall: 0.9999 - auc: 0.9911 - sensitivity: 0.0000e+00 - specificity: 0.9824 - val_loss: 0.2081 - val_tp: 137.0000 - val_fp: 3444.0000 - val_tn: 81851.0000 - val_fn: 11.0000 - val_accuracy: 0.9596 - val_precision: 0.0383 - val_recall: 0.9257 - val_auc: 0.9626 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9819
Epoch 12/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.2937 - tp: 199003.0000 - fp: 12649.0000 - tn: 186371.0000 - fn: 17.0000 - accuracy: 0.9682 - precision: 0.9402 - recall: 0.9999 - auc: 0.9916 - sensitivity: 0.0000e+00 - specificity: 0.9833 - val_loss: 0.1803 - val_tp: 135.0000 - val_fp: 3010.0000 - val_tn: 82285.0000 - val_fn: 13.0000 - val_accuracy: 0.9646 - val_precision: 0.0429 - val_recall: 0.9122 - val_auc: 0.9636 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9862
Epoch 13/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.2915 - tp: 198999.0000 - fp: 12299.0000 - tn: 186721.0000 - fn: 21.0000 - accuracy: 0.9690 - precision: 0.9418 - recall: 0.9999 - auc: 0.9916 - sensitivity: 0.0000e+00 - specificity: 0.9833 - val_loss: 0.1661 - val_tp: 134.0000 - val_fp: 2967.0000 - val_tn: 82328.0000 - val_fn: 14.0000 - val_accuracy: 0.9651 - val_precision: 0.0432 - val_recall: 0.9054 - val_auc: 0.9618 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9844
Epoch 14/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2812 - tp: 199001.0000 - fp: 12043.0000 - tn: 186977.0000 - fn: 19.0000 - accuracy: 0.9697 - precision: 0.9429 - recall: 0.9999 - auc: 0.9921 - sensitivity: 0.0000e+00 - specificity: 0.9843 - val_loss: 0.1882 - val_tp: 137.0000 - val_fp: 3114.0000 - val_tn: 82181.0000 - val_fn: 11.0000 - val_accuracy: 0.9634 - val_precision: 0.0421 - val_recall: 0.9257 - val_auc: 0.9593 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9817
Epoch 15/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2643 - tp: 199003.0000 - fp: 12181.0000 - tn: 186839.0000 - fn: 17.0000 - accuracy: 0.9694 - precision: 0.9423 - recall: 0.9999 - auc: 0.9923 - sensitivity: 0.0000e+00 - specificity: 0.9848 - val_loss: 0.1633 - val_tp: 135.0000 - val_fp: 3182.0000 - val_tn: 82113.0000 - val_fn: 13.0000 - val_accuracy: 0.9626 - val_precision: 0.0407 - val_recall: 0.9122 - val_auc: 0.9618 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9880
Epoch 16/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2985 - tp: 198996.0000 - fp: 12744.0000 - tn: 186276.0000 - fn: 24.0000 - accuracy: 0.9679 - precision: 0.9398 - recall: 0.9999 - auc: 0.9928 - sensitivity: 0.0000e+00 - specificity: 0.9859 - val_loss: 0.1460 - val_tp: 134.0000 - val_fp: 2809.0000 - val_tn: 82486.0000 - val_fn: 14.0000 - val_accuracy: 0.9670 - val_precision: 0.0455 - val_recall: 0.9054 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9892
Epoch 17/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2721 - tp: 198996.0000 - fp: 11764.0000 - tn: 187256.0000 - fn: 24.0000 - accuracy: 0.9704 - precision: 0.9442 - recall: 0.9999 - auc: 0.9933 - sensitivity: 0.0000e+00 - specificity: 0.9867 - val_loss: 0.1669 - val_tp: 135.0000 - val_fp: 3394.0000 - val_tn: 81901.0000 - val_fn: 13.0000 - val_accuracy: 0.9601 - val_precision: 0.0383 - val_recall: 0.9122 - val_auc: 0.9611 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9902
Epoch 18/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2546 - tp: 199001.0000 - fp: 11473.0000 - tn: 187547.0000 - fn: 19.0000 - accuracy: 0.9711 - precision: 0.9455 - recall: 0.9999 - auc: 0.9937 - sensitivity: 0.0000e+00 - specificity: 0.9876 - val_loss: 0.1582 - val_tp: 135.0000 - val_fp: 2877.0000 - val_tn: 82418.0000 - val_fn: 13.0000 - val_accuracy: 0.9662 - val_precision: 0.0448 - val_recall: 0.9122 - val_auc: 0.9611 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9870
Epoch 19/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2480 - tp: 198999.0000 - fp: 12228.0000 - tn: 186792.0000 - fn: 21.0000 - accuracy: 0.9692 - precision: 0.9421 - recall: 0.9999 - auc: 0.9936 - sensitivity: 0.0000e+00 - specificity: 0.9874 - val_loss: 0.1630 - val_tp: 135.0000 - val_fp: 3432.0000 - val_tn: 81863.0000 - val_fn: 13.0000 - val_accuracy: 0.9597 - val_precision: 0.0378 - val_recall: 0.9122 - val_auc: 0.9609 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9887
Epoch 20/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.2298 - tp: 199000.0000 - fp: 11810.0000 - tn: 187210.0000 - fn: 20.0000 - accuracy: 0.9703 - precision: 0.9440 - recall: 0.9999 - auc: 0.9939 - sensitivity: 0.0000e+00 - specificity: 0.9879 - val_loss: 0.1397 - val_tp: 134.0000 - val_fp: 2619.0000 - val_tn: 82676.0000 - val_fn: 14.0000 - val_accuracy: 0.9692 - val_precision: 0.0487 - val_recall: 0.9054 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9868
Epoch 21/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2413 - tp: 199001.0000 - fp: 11891.0000 - tn: 187129.0000 - fn: 19.0000 - accuracy: 0.9701 - precision: 0.9436 - recall: 0.9999 - auc: 0.9942 - sensitivity: 0.0000e+00 - specificity: 0.9884 - val_loss: 0.1363 - val_tp: 135.0000 - val_fp: 2567.0000 - val_tn: 82728.0000 - val_fn: 13.0000 - val_accuracy: 0.9698 - val_precision: 0.0500 - val_recall: 0.9122 - val_auc: 0.9616 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9909
Epoch 22/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.2435 - tp: 198998.0000 - fp: 10942.0000 - tn: 188078.0000 - fn: 22.0000 - accuracy: 0.9725 - precision: 0.9479 - recall: 0.9999 - auc: 0.9949 - sensitivity: 0.0000e+00 - specificity: 0.9898 - val_loss: 0.1313 - val_tp: 133.0000 - val_fp: 2525.0000 - val_tn: 82770.0000 - val_fn: 15.0000 - val_accuracy: 0.9703 - val_precision: 0.0500 - val_recall: 0.8986 - val_auc: 0.9639 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9902
Epoch 23/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1841 - tp: 199001.0000 - fp: 10022.0000 - tn: 188998.0000 - fn: 19.0000 - accuracy: 0.9748 - precision: 0.9521 - recall: 0.9999 - auc: 0.9947 - sensitivity: 0.0000e+00 - specificity: 0.9895 - val_loss: 0.1289 - val_tp: 133.0000 - val_fp: 2451.0000 - val_tn: 82844.0000 - val_fn: 15.0000 - val_accuracy: 0.9711 - val_precision: 0.0515 - val_recall: 0.8986 - val_auc: 0.9645 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9909
Epoch 24/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.2091 - tp: 199001.0000 - fp: 10301.0000 - tn: 188719.0000 - fn: 19.0000 - accuracy: 0.9741 - precision: 0.9508 - recall: 0.9999 - auc: 0.9948 - sensitivity: 0.0000e+00 - specificity: 0.9898 - val_loss: 0.1509 - val_tp: 134.0000 - val_fp: 2748.0000 - val_tn: 82547.0000 - val_fn: 14.0000 - val_accuracy: 0.9677 - val_precision: 0.0465 - val_recall: 0.9054 - val_auc: 0.9613 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9894
Epoch 25/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1924 - tp: 199001.0000 - fp: 10585.0000 - tn: 188435.0000 - fn: 19.0000 - accuracy: 0.9734 - precision: 0.9495 - recall: 0.9999 - auc: 0.9949 - sensitivity: 0.0000e+00 - specificity: 0.9899 - val_loss: 0.1372 - val_tp: 133.0000 - val_fp: 2475.0000 - val_tn: 82820.0000 - val_fn: 15.0000 - val_accuracy: 0.9709 - val_precision: 0.0510 - val_recall: 0.8986 - val_auc: 0.9643 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9917
Epoch 26/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.2143 - tp: 198999.0000 - fp: 10436.0000 - tn: 188584.0000 - fn: 21.0000 - accuracy: 0.9737 - precision: 0.9502 - recall: 0.9999 - auc: 0.9949 - sensitivity: 0.0000e+00 - specificity: 0.9899 - val_loss: 0.1379 - val_tp: 135.0000 - val_fp: 2685.0000 - val_tn: 82610.0000 - val_fn: 13.0000 - val_accuracy: 0.9684 - val_precision: 0.0479 - val_recall: 0.9122 - val_auc: 0.9661 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9925
Epoch 27/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1997 - tp: 199001.0000 - fp: 10362.0000 - tn: 188658.0000 - fn: 19.0000 - accuracy: 0.9739 - precision: 0.9505 - recall: 0.9999 - auc: 0.9951 - sensitivity: 0.0000e+00 - specificity: 0.9903 - val_loss: 0.1230 - val_tp: 133.0000 - val_fp: 2268.0000 - val_tn: 83027.0000 - val_fn: 15.0000 - val_accuracy: 0.9733 - val_precision: 0.0554 - val_recall: 0.8986 - val_auc: 0.9634 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9918
Epoch 28/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1923 - tp: 199000.0000 - fp: 10756.0000 - tn: 188264.0000 - fn: 20.0000 - accuracy: 0.9729 - precision: 0.9487 - recall: 0.9999 - auc: 0.9952 - sensitivity: 0.0000e+00 - specificity: 0.9904 - val_loss: 0.1140 - val_tp: 132.0000 - val_fp: 2003.0000 - val_tn: 83292.0000 - val_fn: 16.0000 - val_accuracy: 0.9764 - val_precision: 0.0618 - val_recall: 0.8919 - val_auc: 0.9606 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9919
Epoch 29/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1716 - tp: 199003.0000 - fp: 9232.0000 - tn: 189788.0000 - fn: 17.0000 - accuracy: 0.9768 - precision: 0.9557 - recall: 0.9999 - auc: 0.9957 - sensitivity: 0.0000e+00 - specificity: 0.9914 - val_loss: 0.1078 - val_tp: 131.0000 - val_fp: 1790.0000 - val_tn: 83505.0000 - val_fn: 17.0000 - val_accuracy: 0.9789 - val_precision: 0.0682 - val_recall: 0.8851 - val_auc: 0.9607 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9918
Epoch 30/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1936 - tp: 199000.0000 - fp: 9156.0000 - tn: 189864.0000 - fn: 20.0000 - accuracy: 0.9769 - precision: 0.9560 - recall: 0.9999 - auc: 0.9955 - sensitivity: 0.0000e+00 - specificity: 0.9910 - val_loss: 0.1093 - val_tp: 132.0000 - val_fp: 1870.0000 - val_tn: 83425.0000 - val_fn: 16.0000 - val_accuracy: 0.9779 - val_precision: 0.0659 - val_recall: 0.8919 - val_auc: 0.9637 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9915
Epoch 31/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1883 - tp: 199002.0000 - fp: 9269.0000 - tn: 189751.0000 - fn: 18.0000 - accuracy: 0.9767 - precision: 0.9555 - recall: 0.9999 - auc: 0.9956 - sensitivity: 0.0000e+00 - specificity: 0.9914 - val_loss: 0.1190 - val_tp: 133.0000 - val_fp: 2157.0000 - val_tn: 83138.0000 - val_fn: 15.0000 - val_accuracy: 0.9746 - val_precision: 0.0581 - val_recall: 0.8986 - val_auc: 0.9647 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9895
Epoch 32/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1897 - tp: 199002.0000 - fp: 9024.0000 - tn: 189996.0000 - fn: 18.0000 - accuracy: 0.9773 - precision: 0.9566 - recall: 0.9999 - auc: 0.9958 - sensitivity: 0.0000e+00 - specificity: 0.9917 - val_loss: 0.1450 - val_tp: 133.0000 - val_fp: 2701.0000 - val_tn: 82594.0000 - val_fn: 15.0000 - val_accuracy: 0.9682 - val_precision: 0.0469 - val_recall: 0.8986 - val_auc: 0.9632 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9916
Epoch 33/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1832 - tp: 199003.0000 - fp: 10066.0000 - tn: 188954.0000 - fn: 17.0000 - accuracy: 0.9747 - precision: 0.9519 - recall: 0.9999 - auc: 0.9960 - sensitivity: 0.0000e+00 - specificity: 0.9920 - val_loss: 0.1138 - val_tp: 131.0000 - val_fp: 1816.0000 - val_tn: 83479.0000 - val_fn: 17.0000 - val_accuracy: 0.9785 - val_precision: 0.0673 - val_recall: 0.8851 - val_auc: 0.9626 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9925
Epoch 34/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1893 - tp: 198999.0000 - fp: 8628.0000 - tn: 190392.0000 - fn: 21.0000 - accuracy: 0.9783 - precision: 0.9584 - recall: 0.9999 - auc: 0.9958 - sensitivity: 0.0000e+00 - specificity: 0.9917 - val_loss: 0.1148 - val_tp: 132.0000 - val_fp: 1782.0000 - val_tn: 83513.0000 - val_fn: 16.0000 - val_accuracy: 0.9790 - val_precision: 0.0690 - val_recall: 0.8919 - val_auc: 0.9642 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9905
Epoch 35/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1711 - tp: 199000.0000 - fp: 8653.0000 - tn: 190367.0000 - fn: 20.0000 - accuracy: 0.9782 - precision: 0.9583 - recall: 0.9999 - auc: 0.9960 - sensitivity: 0.0000e+00 - specificity: 0.9920 - val_loss: 0.1129 - val_tp: 132.0000 - val_fp: 1962.0000 - val_tn: 83333.0000 - val_fn: 16.0000 - val_accuracy: 0.9769 - val_precision: 0.0630 - val_recall: 0.8919 - val_auc: 0.9643 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9912
Epoch 36/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1715 - tp: 199000.0000 - fp: 8061.0000 - tn: 190959.0000 - fn: 20.0000 - accuracy: 0.9797 - precision: 0.9611 - recall: 0.9999 - auc: 0.9963 - sensitivity: 0.0000e+00 - specificity: 0.9926 - val_loss: 0.1304 - val_tp: 133.0000 - val_fp: 2320.0000 - val_tn: 82975.0000 - val_fn: 15.0000 - val_accuracy: 0.9727 - val_precision: 0.0542 - val_recall: 0.8986 - val_auc: 0.9644 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9928
Epoch 37/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1860 - tp: 199002.0000 - fp: 9671.0000 - tn: 189349.0000 - fn: 18.0000 - accuracy: 0.9757 - precision: 0.9537 - recall: 0.9999 - auc: 0.9958 - sensitivity: 0.0000e+00 - specificity: 0.9917 - val_loss: 0.1218 - val_tp: 133.0000 - val_fp: 2373.0000 - val_tn: 82922.0000 - val_fn: 15.0000 - val_accuracy: 0.9721 - val_precision: 0.0531 - val_recall: 0.8986 - val_auc: 0.9631 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9934
Epoch 38/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1695 - tp: 199000.0000 - fp: 9800.0000 - tn: 189220.0000 - fn: 20.0000 - accuracy: 0.9753 - precision: 0.9531 - recall: 0.9999 - auc: 0.9959 - sensitivity: 0.0000e+00 - specificity: 0.9919 - val_loss: 0.1122 - val_tp: 133.0000 - val_fp: 1892.0000 - val_tn: 83403.0000 - val_fn: 15.0000 - val_accuracy: 0.9777 - val_precision: 0.0657 - val_recall: 0.8986 - val_auc: 0.9637 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 39/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1993 - tp: 199000.0000 - fp: 8309.0000 - tn: 190711.0000 - fn: 20.0000 - accuracy: 0.9791 - precision: 0.9599 - recall: 0.9999 - auc: 0.9962 - sensitivity: 0.0000e+00 - specificity: 0.9924 - val_loss: 0.1335 - val_tp: 133.0000 - val_fp: 2436.0000 - val_tn: 82859.0000 - val_fn: 15.0000 - val_accuracy: 0.9713 - val_precision: 0.0518 - val_recall: 0.8986 - val_auc: 0.9626 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 40/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1773 - tp: 199001.0000 - fp: 9365.0000 - tn: 189655.0000 - fn: 19.0000 - accuracy: 0.9764 - precision: 0.9551 - recall: 0.9999 - auc: 0.9961 - sensitivity: 0.0000e+00 - specificity: 0.9923 - val_loss: 0.1108 - val_tp: 132.0000 - val_fp: 1758.0000 - val_tn: 83537.0000 - val_fn: 16.0000 - val_accuracy: 0.9792 - val_precision: 0.0698 - val_recall: 0.8919 - val_auc: 0.9645 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9910
Epoch 41/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1562 - tp: 199003.0000 - fp: 7118.0000 - tn: 191902.0000 - fn: 17.0000 - accuracy: 0.9821 - precision: 0.9655 - recall: 0.9999 - auc: 0.9966 - sensitivity: 0.0000e+00 - specificity: 0.9933 - val_loss: 0.1049 - val_tp: 131.0000 - val_fp: 1820.0000 - val_tn: 83475.0000 - val_fn: 17.0000 - val_accuracy: 0.9785 - val_precision: 0.0671 - val_recall: 0.8851 - val_auc: 0.9638 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 42/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1572 - tp: 199003.0000 - fp: 7806.0000 - tn: 191214.0000 - fn: 17.0000 - accuracy: 0.9803 - precision: 0.9623 - recall: 0.9999 - auc: 0.9962 - sensitivity: 0.0000e+00 - specificity: 0.9924 - val_loss: 0.1111 - val_tp: 131.0000 - val_fp: 1820.0000 - val_tn: 83475.0000 - val_fn: 17.0000 - val_accuracy: 0.9785 - val_precision: 0.0671 - val_recall: 0.8851 - val_auc: 0.9631 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9902
Epoch 43/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1624 - tp: 199001.0000 - fp: 8848.0000 - tn: 190172.0000 - fn: 19.0000 - accuracy: 0.9777 - precision: 0.9574 - recall: 0.9999 - auc: 0.9962 - sensitivity: 0.0000e+00 - specificity: 0.9925 - val_loss: 0.0927 - val_tp: 132.0000 - val_fp: 1511.0000 - val_tn: 83784.0000 - val_fn: 16.0000 - val_accuracy: 0.9821 - val_precision: 0.0803 - val_recall: 0.8919 - val_auc: 0.9654 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9919
Epoch 44/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1820 - tp: 198999.0000 - fp: 8520.0000 - tn: 190500.0000 - fn: 21.0000 - accuracy: 0.9785 - precision: 0.9589 - recall: 0.9999 - auc: 0.9961 - sensitivity: 0.0000e+00 - specificity: 0.9922 - val_loss: 0.1015 - val_tp: 133.0000 - val_fp: 1832.0000 - val_tn: 83463.0000 - val_fn: 15.0000 - val_accuracy: 0.9784 - val_precision: 0.0677 - val_recall: 0.8986 - val_auc: 0.9630 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9932
Epoch 45/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1878 - tp: 198997.0000 - fp: 9108.0000 - tn: 189912.0000 - fn: 23.0000 - accuracy: 0.9771 - precision: 0.9562 - recall: 0.9999 - auc: 0.9964 - sensitivity: 0.0000e+00 - specificity: 0.9928 - val_loss: 0.1102 - val_tp: 133.0000 - val_fp: 1881.0000 - val_tn: 83414.0000 - val_fn: 15.0000 - val_accuracy: 0.9778 - val_precision: 0.0660 - val_recall: 0.8986 - val_auc: 0.9650 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9931
Epoch 46/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1783 - tp: 198998.0000 - fp: 8940.0000 - tn: 190080.0000 - fn: 22.0000 - accuracy: 0.9775 - precision: 0.9570 - recall: 0.9999 - auc: 0.9965 - sensitivity: 0.0000e+00 - specificity: 0.9931 - val_loss: 0.0955 - val_tp: 133.0000 - val_fp: 1722.0000 - val_tn: 83573.0000 - val_fn: 15.0000 - val_accuracy: 0.9797 - val_precision: 0.0717 - val_recall: 0.8986 - val_auc: 0.9651 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9919
Epoch 47/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1584 - tp: 199004.0000 - fp: 8121.0000 - tn: 190899.0000 - fn: 16.0000 - accuracy: 0.9796 - precision: 0.9608 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9935 - val_loss: 0.1037 - val_tp: 133.0000 - val_fp: 1902.0000 - val_tn: 83393.0000 - val_fn: 15.0000 - val_accuracy: 0.9776 - val_precision: 0.0654 - val_recall: 0.8986 - val_auc: 0.9610 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9925
Epoch 48/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1498 - tp: 199003.0000 - fp: 7932.0000 - tn: 191088.0000 - fn: 17.0000 - accuracy: 0.9800 - precision: 0.9617 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9934 - val_loss: 0.0918 - val_tp: 133.0000 - val_fp: 1618.0000 - val_tn: 83677.0000 - val_fn: 15.0000 - val_accuracy: 0.9809 - val_precision: 0.0760 - val_recall: 0.8986 - val_auc: 0.9637 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9932
Epoch 49/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1566 - tp: 199002.0000 - fp: 8293.0000 - tn: 190727.0000 - fn: 18.0000 - accuracy: 0.9791 - precision: 0.9600 - recall: 0.9999 - auc: 0.9965 - sensitivity: 0.0000e+00 - specificity: 0.9930 - val_loss: 0.0966 - val_tp: 132.0000 - val_fp: 1553.0000 - val_tn: 83742.0000 - val_fn: 16.0000 - val_accuracy: 0.9816 - val_precision: 0.0783 - val_recall: 0.8919 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 50/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1745 - tp: 198998.0000 - fp: 8141.0000 - tn: 190879.0000 - fn: 22.0000 - accuracy: 0.9795 - precision: 0.9607 - recall: 0.9999 - auc: 0.9966 - sensitivity: 0.0000e+00 - specificity: 0.9933 - val_loss: 0.1090 - val_tp: 133.0000 - val_fp: 1973.0000 - val_tn: 83322.0000 - val_fn: 15.0000 - val_accuracy: 0.9767 - val_precision: 0.0632 - val_recall: 0.8986 - val_auc: 0.9643 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 51/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1513 - tp: 199003.0000 - fp: 9172.0000 - tn: 189848.0000 - fn: 17.0000 - accuracy: 0.9769 - precision: 0.9559 - recall: 0.9999 - auc: 0.9965 - sensitivity: 0.0000e+00 - specificity: 0.9931 - val_loss: 0.1158 - val_tp: 132.0000 - val_fp: 2171.0000 - val_tn: 83124.0000 - val_fn: 16.0000 - val_accuracy: 0.9744 - val_precision: 0.0573 - val_recall: 0.8919 - val_auc: 0.9645 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 52/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1570 - tp: 199000.0000 - fp: 8634.0000 - tn: 190386.0000 - fn: 20.0000 - accuracy: 0.9783 - precision: 0.9584 - recall: 0.9999 - auc: 0.9966 - sensitivity: 0.0000e+00 - specificity: 0.9933 - val_loss: 0.1043 - val_tp: 133.0000 - val_fp: 1921.0000 - val_tn: 83374.0000 - val_fn: 15.0000 - val_accuracy: 0.9773 - val_precision: 0.0648 - val_recall: 0.8986 - val_auc: 0.9633 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9909
Epoch 53/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1624 - tp: 199001.0000 - fp: 8124.0000 - tn: 190896.0000 - fn: 19.0000 - accuracy: 0.9795 - precision: 0.9608 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9934 - val_loss: 0.1129 - val_tp: 134.0000 - val_fp: 2180.0000 - val_tn: 83115.0000 - val_fn: 14.0000 - val_accuracy: 0.9743 - val_precision: 0.0579 - val_recall: 0.9054 - val_auc: 0.9665 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9908
Epoch 54/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1555 - tp: 199000.0000 - fp: 7626.0000 - tn: 191394.0000 - fn: 20.0000 - accuracy: 0.9808 - precision: 0.9631 - recall: 0.9999 - auc: 0.9964 - sensitivity: 0.0000e+00 - specificity: 0.9929 - val_loss: 0.1201 - val_tp: 134.0000 - val_fp: 2279.0000 - val_tn: 83016.0000 - val_fn: 14.0000 - val_accuracy: 0.9732 - val_precision: 0.0555 - val_recall: 0.9054 - val_auc: 0.9679 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 55/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1524 - tp: 199003.0000 - fp: 8774.0000 - tn: 190246.0000 - fn: 17.0000 - accuracy: 0.9779 - precision: 0.9578 - recall: 0.9999 - auc: 0.9968 - sensitivity: 0.0000e+00 - specificity: 0.9936 - val_loss: 0.0983 - val_tp: 132.0000 - val_fp: 1824.0000 - val_tn: 83471.0000 - val_fn: 16.0000 - val_accuracy: 0.9785 - val_precision: 0.0675 - val_recall: 0.8919 - val_auc: 0.9635 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9938
Epoch 56/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1642 - tp: 199000.0000 - fp: 7970.0000 - tn: 191050.0000 - fn: 20.0000 - accuracy: 0.9799 - precision: 0.9615 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9934 - val_loss: 0.1042 - val_tp: 132.0000 - val_fp: 1965.0000 - val_tn: 83330.0000 - val_fn: 16.0000 - val_accuracy: 0.9768 - val_precision: 0.0629 - val_recall: 0.8919 - val_auc: 0.9628 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9930
Epoch 57/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1491 - tp: 199001.0000 - fp: 8386.0000 - tn: 190634.0000 - fn: 19.0000 - accuracy: 0.9789 - precision: 0.9596 - recall: 0.9999 - auc: 0.9968 - sensitivity: 0.0000e+00 - specificity: 0.9936 - val_loss: 0.0854 - val_tp: 132.0000 - val_fp: 1441.0000 - val_tn: 83854.0000 - val_fn: 16.0000 - val_accuracy: 0.9829 - val_precision: 0.0839 - val_recall: 0.8919 - val_auc: 0.9643 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9917
Epoch 58/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1327 - tp: 199004.0000 - fp: 6788.0000 - tn: 192232.0000 - fn: 16.0000 - accuracy: 0.9829 - precision: 0.9670 - recall: 0.9999 - auc: 0.9966 - sensitivity: 0.0000e+00 - specificity: 0.9933 - val_loss: 0.1044 - val_tp: 132.0000 - val_fp: 1706.0000 - val_tn: 83589.0000 - val_fn: 16.0000 - val_accuracy: 0.9798 - val_precision: 0.0718 - val_recall: 0.8919 - val_auc: 0.9628 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9913
Epoch 59/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1388 - tp: 199000.0000 - fp: 6551.0000 - tn: 192469.0000 - fn: 20.0000 - accuracy: 0.9835 - precision: 0.9681 - recall: 0.9999 - auc: 0.9965 - sensitivity: 0.0000e+00 - specificity: 0.9931 - val_loss: 0.1120 - val_tp: 134.0000 - val_fp: 2034.0000 - val_tn: 83261.0000 - val_fn: 14.0000 - val_accuracy: 0.9760 - val_precision: 0.0618 - val_recall: 0.9054 - val_auc: 0.9663 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9930
Epoch 60/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1519 - tp: 199002.0000 - fp: 8644.0000 - tn: 190376.0000 - fn: 18.0000 - accuracy: 0.9782 - precision: 0.9584 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9934 - val_loss: 0.1016 - val_tp: 132.0000 - val_fp: 1714.0000 - val_tn: 83581.0000 - val_fn: 16.0000 - val_accuracy: 0.9798 - val_precision: 0.0715 - val_recall: 0.8919 - val_auc: 0.9652 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9913
Epoch 61/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1616 - tp: 198999.0000 - fp: 7989.0000 - tn: 191031.0000 - fn: 21.0000 - accuracy: 0.9799 - precision: 0.9614 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9934 - val_loss: 0.0880 - val_tp: 131.0000 - val_fp: 1471.0000 - val_tn: 83824.0000 - val_fn: 17.0000 - val_accuracy: 0.9826 - val_precision: 0.0818 - val_recall: 0.8851 - val_auc: 0.9644 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9915
Epoch 62/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1431 - tp: 199004.0000 - fp: 7489.0000 - tn: 191531.0000 - fn: 16.0000 - accuracy: 0.9811 - precision: 0.9637 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9938 - val_loss: 0.0891 - val_tp: 131.0000 - val_fp: 1481.0000 - val_tn: 83814.0000 - val_fn: 17.0000 - val_accuracy: 0.9825 - val_precision: 0.0813 - val_recall: 0.8851 - val_auc: 0.9587 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9914
Epoch 63/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1483 - tp: 199002.0000 - fp: 7742.0000 - tn: 191278.0000 - fn: 18.0000 - accuracy: 0.9805 - precision: 0.9626 - recall: 0.9999 - auc: 0.9968 - sensitivity: 0.0000e+00 - specificity: 0.9936 - val_loss: 0.0875 - val_tp: 132.0000 - val_fp: 1543.0000 - val_tn: 83752.0000 - val_fn: 16.0000 - val_accuracy: 0.9818 - val_precision: 0.0788 - val_recall: 0.8919 - val_auc: 0.9608 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 64/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1669 - tp: 199001.0000 - fp: 8469.0000 - tn: 190551.0000 - fn: 19.0000 - accuracy: 0.9787 - precision: 0.9592 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9941 - val_loss: 0.0862 - val_tp: 131.0000 - val_fp: 1555.0000 - val_tn: 83740.0000 - val_fn: 17.0000 - val_accuracy: 0.9816 - val_precision: 0.0777 - val_recall: 0.8851 - val_auc: 0.9571 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 65/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1466 - tp: 199003.0000 - fp: 7565.0000 - tn: 191455.0000 - fn: 17.0000 - accuracy: 0.9810 - precision: 0.9634 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9939 - val_loss: 0.1066 - val_tp: 132.0000 - val_fp: 1971.0000 - val_tn: 83324.0000 - val_fn: 16.0000 - val_accuracy: 0.9767 - val_precision: 0.0628 - val_recall: 0.8919 - val_auc: 0.9601 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9905
Epoch 66/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1410 - tp: 199005.0000 - fp: 8330.0000 - tn: 190690.0000 - fn: 15.0000 - accuracy: 0.9790 - precision: 0.9598 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9938 - val_loss: 0.0914 - val_tp: 131.0000 - val_fp: 1538.0000 - val_tn: 83757.0000 - val_fn: 17.0000 - val_accuracy: 0.9818 - val_precision: 0.0785 - val_recall: 0.8851 - val_auc: 0.9609 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9917
Epoch 67/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1493 - tp: 199000.0000 - fp: 7600.0000 - tn: 191420.0000 - fn: 20.0000 - accuracy: 0.9809 - precision: 0.9632 - recall: 0.9999 - auc: 0.9967 - sensitivity: 0.0000e+00 - specificity: 0.9935 - val_loss: 0.1280 - val_tp: 135.0000 - val_fp: 2175.0000 - val_tn: 83120.0000 - val_fn: 13.0000 - val_accuracy: 0.9744 - val_precision: 0.0584 - val_recall: 0.9122 - val_auc: 0.9592 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9924
Epoch 68/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1508 - tp: 199000.0000 - fp: 8819.0000 - tn: 190201.0000 - fn: 20.0000 - accuracy: 0.9778 - precision: 0.9576 - recall: 0.9999 - auc: 0.9966 - sensitivity: 0.0000e+00 - specificity: 0.9932 - val_loss: 0.1036 - val_tp: 131.0000 - val_fp: 1809.0000 - val_tn: 83486.0000 - val_fn: 17.0000 - val_accuracy: 0.9786 - val_precision: 0.0675 - val_recall: 0.8851 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9943
Epoch 69/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1373 - tp: 199002.0000 - fp: 7571.0000 - tn: 191449.0000 - fn: 18.0000 - accuracy: 0.9809 - precision: 0.9633 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9939 - val_loss: 0.1026 - val_tp: 133.0000 - val_fp: 1933.0000 - val_tn: 83362.0000 - val_fn: 15.0000 - val_accuracy: 0.9772 - val_precision: 0.0644 - val_recall: 0.8986 - val_auc: 0.9604 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9924
Epoch 70/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1453 - tp: 199000.0000 - fp: 8230.0000 - tn: 190790.0000 - fn: 20.0000 - accuracy: 0.9793 - precision: 0.9603 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9940 - val_loss: 0.1041 - val_tp: 132.0000 - val_fp: 1898.0000 - val_tn: 83397.0000 - val_fn: 16.0000 - val_accuracy: 0.9776 - val_precision: 0.0650 - val_recall: 0.8919 - val_auc: 0.9577 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9898
Epoch 71/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1278 - tp: 199006.0000 - fp: 7255.0000 - tn: 191765.0000 - fn: 14.0000 - accuracy: 0.9817 - precision: 0.9648 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9941 - val_loss: 0.0918 - val_tp: 132.0000 - val_fp: 1622.0000 - val_tn: 83673.0000 - val_fn: 16.0000 - val_accuracy: 0.9808 - val_precision: 0.0753 - val_recall: 0.8919 - val_auc: 0.9580 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9911
Epoch 72/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1340 - tp: 199004.0000 - fp: 7512.0000 - tn: 191508.0000 - fn: 16.0000 - accuracy: 0.9811 - precision: 0.9636 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9939 - val_loss: 0.0798 - val_tp: 130.0000 - val_fp: 1284.0000 - val_tn: 84011.0000 - val_fn: 18.0000 - val_accuracy: 0.9848 - val_precision: 0.0919 - val_recall: 0.8784 - val_auc: 0.9593 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9913
Epoch 73/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1364 - tp: 199004.0000 - fp: 7071.0000 - tn: 191949.0000 - fn: 16.0000 - accuracy: 0.9822 - precision: 0.9657 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9941 - val_loss: 0.0960 - val_tp: 131.0000 - val_fp: 1666.0000 - val_tn: 83629.0000 - val_fn: 17.0000 - val_accuracy: 0.9803 - val_precision: 0.0729 - val_recall: 0.8851 - val_auc: 0.9586 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9901
Epoch 74/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1454 - tp: 199000.0000 - fp: 6920.0000 - tn: 192100.0000 - fn: 20.0000 - accuracy: 0.9826 - precision: 0.9664 - recall: 0.9999 - auc: 0.9971 - sensitivity: 0.0000e+00 - specificity: 0.9942 - val_loss: 0.0937 - val_tp: 133.0000 - val_fp: 1781.0000 - val_tn: 83514.0000 - val_fn: 15.0000 - val_accuracy: 0.9790 - val_precision: 0.0695 - val_recall: 0.8986 - val_auc: 0.9618 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9927
Epoch 75/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1584 - tp: 199004.0000 - fp: 7800.0000 - tn: 191220.0000 - fn: 16.0000 - accuracy: 0.9804 - precision: 0.9623 - recall: 0.9999 - auc: 0.9968 - sensitivity: 0.0000e+00 - specificity: 0.9937 - val_loss: 0.1085 - val_tp: 134.0000 - val_fp: 2252.0000 - val_tn: 83043.0000 - val_fn: 14.0000 - val_accuracy: 0.9735 - val_precision: 0.0562 - val_recall: 0.9054 - val_auc: 0.9624 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9904
Epoch 76/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1388 - tp: 199001.0000 - fp: 8921.0000 - tn: 190099.0000 - fn: 19.0000 - accuracy: 0.9775 - precision: 0.9571 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9937 - val_loss: 0.0845 - val_tp: 131.0000 - val_fp: 1521.0000 - val_tn: 83774.0000 - val_fn: 17.0000 - val_accuracy: 0.9820 - val_precision: 0.0793 - val_recall: 0.8851 - val_auc: 0.9593 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9929
Epoch 77/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1364 - tp: 199006.0000 - fp: 6987.0000 - tn: 192033.0000 - fn: 14.0000 - accuracy: 0.9824 - precision: 0.9661 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9944 - val_loss: 0.0870 - val_tp: 131.0000 - val_fp: 1568.0000 - val_tn: 83727.0000 - val_fn: 17.0000 - val_accuracy: 0.9814 - val_precision: 0.0771 - val_recall: 0.8851 - val_auc: 0.9583 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9955
Epoch 78/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1220 - tp: 199004.0000 - fp: 6681.0000 - tn: 192339.0000 - fn: 16.0000 - accuracy: 0.9832 - precision: 0.9675 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0885 - val_tp: 132.0000 - val_fp: 1588.0000 - val_tn: 83707.0000 - val_fn: 16.0000 - val_accuracy: 0.9812 - val_precision: 0.0767 - val_recall: 0.8919 - val_auc: 0.9616 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9917
Epoch 79/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1418 - tp: 199002.0000 - fp: 8370.0000 - tn: 190650.0000 - fn: 18.0000 - accuracy: 0.9789 - precision: 0.9596 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9939 - val_loss: 0.1048 - val_tp: 132.0000 - val_fp: 1786.0000 - val_tn: 83509.0000 - val_fn: 16.0000 - val_accuracy: 0.9789 - val_precision: 0.0688 - val_recall: 0.8919 - val_auc: 0.9591 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9919
Epoch 80/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1389 - tp: 199002.0000 - fp: 7351.0000 - tn: 191669.0000 - fn: 18.0000 - accuracy: 0.9815 - precision: 0.9644 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9939 - val_loss: 0.0830 - val_tp: 131.0000 - val_fp: 1415.0000 - val_tn: 83880.0000 - val_fn: 17.0000 - val_accuracy: 0.9832 - val_precision: 0.0847 - val_recall: 0.8851 - val_auc: 0.9617 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9913
Epoch 81/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1297 - tp: 199004.0000 - fp: 7558.0000 - tn: 191462.0000 - fn: 16.0000 - accuracy: 0.9810 - precision: 0.9634 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0862 - val_tp: 132.0000 - val_fp: 1557.0000 - val_tn: 83738.0000 - val_fn: 16.0000 - val_accuracy: 0.9816 - val_precision: 0.0782 - val_recall: 0.8919 - val_auc: 0.9620 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9909
Epoch 82/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1477 - tp: 199003.0000 - fp: 7992.0000 - tn: 191028.0000 - fn: 17.0000 - accuracy: 0.9799 - precision: 0.9614 - recall: 0.9999 - auc: 0.9966 - sensitivity: 0.0000e+00 - specificity: 0.9933 - val_loss: 0.0981 - val_tp: 132.0000 - val_fp: 1971.0000 - val_tn: 83324.0000 - val_fn: 16.0000 - val_accuracy: 0.9767 - val_precision: 0.0628 - val_recall: 0.8919 - val_auc: 0.9616 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9913
Epoch 83/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1339 - tp: 199006.0000 - fp: 7594.0000 - tn: 191426.0000 - fn: 14.0000 - accuracy: 0.9809 - precision: 0.9632 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0884 - val_tp: 131.0000 - val_fp: 1718.0000 - val_tn: 83577.0000 - val_fn: 17.0000 - val_accuracy: 0.9797 - val_precision: 0.0708 - val_recall: 0.8851 - val_auc: 0.9620 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9924
Epoch 84/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1374 - tp: 199006.0000 - fp: 7321.0000 - tn: 191699.0000 - fn: 14.0000 - accuracy: 0.9816 - precision: 0.9645 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0839 - val_tp: 131.0000 - val_fp: 1455.0000 - val_tn: 83840.0000 - val_fn: 17.0000 - val_accuracy: 0.9828 - val_precision: 0.0826 - val_recall: 0.8851 - val_auc: 0.9613 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9924
Epoch 85/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1300 - tp: 198999.0000 - fp: 7631.0000 - tn: 191389.0000 - fn: 21.0000 - accuracy: 0.9808 - precision: 0.9631 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.1017 - val_tp: 134.0000 - val_fp: 1972.0000 - val_tn: 83323.0000 - val_fn: 14.0000 - val_accuracy: 0.9768 - val_precision: 0.0636 - val_recall: 0.9054 - val_auc: 0.9615 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9916
Epoch 86/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1318 - tp: 199003.0000 - fp: 7724.0000 - tn: 191296.0000 - fn: 17.0000 - accuracy: 0.9806 - precision: 0.9626 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0804 - val_tp: 131.0000 - val_fp: 1407.0000 - val_tn: 83888.0000 - val_fn: 17.0000 - val_accuracy: 0.9833 - val_precision: 0.0852 - val_recall: 0.8851 - val_auc: 0.9619 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9950
Epoch 87/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1292 - tp: 199003.0000 - fp: 7554.0000 - tn: 191466.0000 - fn: 17.0000 - accuracy: 0.9810 - precision: 0.9634 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9947 - val_loss: 0.0822 - val_tp: 129.0000 - val_fp: 1387.0000 - val_tn: 83908.0000 - val_fn: 19.0000 - val_accuracy: 0.9835 - val_precision: 0.0851 - val_recall: 0.8716 - val_auc: 0.9643 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9955
Epoch 88/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1253 - tp: 199004.0000 - fp: 6992.0000 - tn: 192028.0000 - fn: 16.0000 - accuracy: 0.9824 - precision: 0.9661 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0896 - val_tp: 129.0000 - val_fp: 1508.0000 - val_tn: 83787.0000 - val_fn: 19.0000 - val_accuracy: 0.9821 - val_precision: 0.0788 - val_recall: 0.8716 - val_auc: 0.9595 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9907
Epoch 89/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1122 - tp: 199005.0000 - fp: 6737.0000 - tn: 192283.0000 - fn: 15.0000 - accuracy: 0.9830 - precision: 0.9673 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0804 - val_tp: 130.0000 - val_fp: 1424.0000 - val_tn: 83871.0000 - val_fn: 18.0000 - val_accuracy: 0.9831 - val_precision: 0.0837 - val_recall: 0.8784 - val_auc: 0.9596 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9902
Epoch 90/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1152 - tp: 199005.0000 - fp: 6937.0000 - tn: 192083.0000 - fn: 15.0000 - accuracy: 0.9825 - precision: 0.9663 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.1052 - val_tp: 131.0000 - val_fp: 1885.0000 - val_tn: 83410.0000 - val_fn: 17.0000 - val_accuracy: 0.9777 - val_precision: 0.0650 - val_recall: 0.8851 - val_auc: 0.9617 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9917
Epoch 91/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1248 - tp: 199004.0000 - fp: 7987.0000 - tn: 191033.0000 - fn: 16.0000 - accuracy: 0.9799 - precision: 0.9614 - recall: 0.9999 - auc: 0.9971 - sensitivity: 0.0000e+00 - specificity: 0.9942 - val_loss: 0.0938 - val_tp: 130.0000 - val_fp: 1704.0000 - val_tn: 83591.0000 - val_fn: 18.0000 - val_accuracy: 0.9798 - val_precision: 0.0709 - val_recall: 0.8784 - val_auc: 0.9587 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9906
Epoch 92/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1252 - tp: 199004.0000 - fp: 7045.0000 - tn: 191975.0000 - fn: 16.0000 - accuracy: 0.9823 - precision: 0.9658 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9947 - val_loss: 0.0878 - val_tp: 130.0000 - val_fp: 1574.0000 - val_tn: 83721.0000 - val_fn: 18.0000 - val_accuracy: 0.9814 - val_precision: 0.0763 - val_recall: 0.8784 - val_auc: 0.9615 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9920
Epoch 93/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1478 - tp: 199001.0000 - fp: 7735.0000 - tn: 191285.0000 - fn: 19.0000 - accuracy: 0.9805 - precision: 0.9626 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0848 - val_tp: 131.0000 - val_fp: 1572.0000 - val_tn: 83723.0000 - val_fn: 17.0000 - val_accuracy: 0.9814 - val_precision: 0.0769 - val_recall: 0.8851 - val_auc: 0.9636 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9917
Epoch 94/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1320 - tp: 199003.0000 - fp: 6820.0000 - tn: 192200.0000 - fn: 17.0000 - accuracy: 0.9828 - precision: 0.9669 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9940 - val_loss: 0.1032 - val_tp: 131.0000 - val_fp: 1857.0000 - val_tn: 83438.0000 - val_fn: 17.0000 - val_accuracy: 0.9781 - val_precision: 0.0659 - val_recall: 0.8851 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9894
Epoch 95/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1278 - tp: 199003.0000 - fp: 7125.0000 - tn: 191895.0000 - fn: 17.0000 - accuracy: 0.9821 - precision: 0.9654 - recall: 0.9999 - auc: 0.9971 - sensitivity: 0.0000e+00 - specificity: 0.9942 - val_loss: 0.0771 - val_tp: 130.0000 - val_fp: 1415.0000 - val_tn: 83880.0000 - val_fn: 18.0000 - val_accuracy: 0.9832 - val_precision: 0.0841 - val_recall: 0.8784 - val_auc: 0.9593 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9932
Epoch 96/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1262 - tp: 199000.0000 - fp: 6971.0000 - tn: 192049.0000 - fn: 20.0000 - accuracy: 0.9824 - precision: 0.9662 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9944 - val_loss: 0.0912 - val_tp: 132.0000 - val_fp: 1788.0000 - val_tn: 83507.0000 - val_fn: 16.0000 - val_accuracy: 0.9789 - val_precision: 0.0688 - val_recall: 0.8919 - val_auc: 0.9616 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9919
Epoch 97/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1161 - tp: 199003.0000 - fp: 7322.0000 - tn: 191698.0000 - fn: 17.0000 - accuracy: 0.9816 - precision: 0.9645 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0882 - val_tp: 132.0000 - val_fp: 1574.0000 - val_tn: 83721.0000 - val_fn: 16.0000 - val_accuracy: 0.9814 - val_precision: 0.0774 - val_recall: 0.8919 - val_auc: 0.9597 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9921
Epoch 98/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1337 - tp: 199004.0000 - fp: 7354.0000 - tn: 191666.0000 - fn: 16.0000 - accuracy: 0.9815 - precision: 0.9644 - recall: 0.9999 - auc: 0.9971 - sensitivity: 0.0000e+00 - specificity: 0.9943 - val_loss: 0.0861 - val_tp: 132.0000 - val_fp: 1674.0000 - val_tn: 83621.0000 - val_fn: 16.0000 - val_accuracy: 0.9802 - val_precision: 0.0731 - val_recall: 0.8919 - val_auc: 0.9592 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 99/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1319 - tp: 199002.0000 - fp: 7304.0000 - tn: 191716.0000 - fn: 18.0000 - accuracy: 0.9816 - precision: 0.9646 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9944 - val_loss: 0.0918 - val_tp: 133.0000 - val_fp: 1735.0000 - val_tn: 83560.0000 - val_fn: 15.0000 - val_accuracy: 0.9795 - val_precision: 0.0712 - val_recall: 0.8986 - val_auc: 0.9594 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9921
Epoch 100/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1313 - tp: 199004.0000 - fp: 7029.0000 - tn: 191991.0000 - fn: 16.0000 - accuracy: 0.9823 - precision: 0.9659 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9948 - val_loss: 0.0832 - val_tp: 131.0000 - val_fp: 1572.0000 - val_tn: 83723.0000 - val_fn: 17.0000 - val_accuracy: 0.9814 - val_precision: 0.0769 - val_recall: 0.8851 - val_auc: 0.9598 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9933
Epoch 101/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1169 - tp: 199007.0000 - fp: 6310.0000 - tn: 192710.0000 - fn: 13.0000 - accuracy: 0.9841 - precision: 0.9693 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0782 - val_tp: 130.0000 - val_fp: 1466.0000 - val_tn: 83829.0000 - val_fn: 18.0000 - val_accuracy: 0.9826 - val_precision: 0.0815 - val_recall: 0.8784 - val_auc: 0.9627 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9960
Epoch 102/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1191 - tp: 199008.0000 - fp: 6526.0000 - tn: 192494.0000 - fn: 12.0000 - accuracy: 0.9836 - precision: 0.9682 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9940 - val_loss: 0.0886 - val_tp: 132.0000 - val_fp: 1571.0000 - val_tn: 83724.0000 - val_fn: 16.0000 - val_accuracy: 0.9814 - val_precision: 0.0775 - val_recall: 0.8919 - val_auc: 0.9623 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9928
Epoch 103/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1189 - tp: 199008.0000 - fp: 7452.0000 - tn: 191568.0000 - fn: 12.0000 - accuracy: 0.9812 - precision: 0.9639 - recall: 0.9999 - auc: 0.9970 - sensitivity: 0.0000e+00 - specificity: 0.9941 - val_loss: 0.0786 - val_tp: 130.0000 - val_fp: 1295.0000 - val_tn: 84000.0000 - val_fn: 18.0000 - val_accuracy: 0.9846 - val_precision: 0.0912 - val_recall: 0.8784 - val_auc: 0.9605 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9932
Epoch 104/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1229 - tp: 199005.0000 - fp: 6312.0000 - tn: 192708.0000 - fn: 15.0000 - accuracy: 0.9841 - precision: 0.9693 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9947 - val_loss: 0.0760 - val_tp: 130.0000 - val_fp: 1351.0000 - val_tn: 83944.0000 - val_fn: 18.0000 - val_accuracy: 0.9840 - val_precision: 0.0878 - val_recall: 0.8784 - val_auc: 0.9597 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9943
Epoch 105/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1426 - tp: 199000.0000 - fp: 7774.0000 - tn: 191246.0000 - fn: 20.0000 - accuracy: 0.9804 - precision: 0.9624 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0868 - val_tp: 132.0000 - val_fp: 1670.0000 - val_tn: 83625.0000 - val_fn: 16.0000 - val_accuracy: 0.9803 - val_precision: 0.0733 - val_recall: 0.8919 - val_auc: 0.9590 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 106/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1133 - tp: 199011.0000 - fp: 7188.0000 - tn: 191832.0000 - fn: 9.0000 - accuracy: 0.9819 - precision: 0.9651 - recall: 1.0000 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0788 - val_tp: 130.0000 - val_fp: 1357.0000 - val_tn: 83938.0000 - val_fn: 18.0000 - val_accuracy: 0.9839 - val_precision: 0.0874 - val_recall: 0.8784 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9939
Epoch 107/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1328 - tp: 198999.0000 - fp: 7003.0000 - tn: 192017.0000 - fn: 21.0000 - accuracy: 0.9824 - precision: 0.9660 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0833 - val_tp: 132.0000 - val_fp: 1700.0000 - val_tn: 83595.0000 - val_fn: 16.0000 - val_accuracy: 0.9799 - val_precision: 0.0721 - val_recall: 0.8919 - val_auc: 0.9599 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9951
Epoch 108/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1253 - tp: 199005.0000 - fp: 7127.0000 - tn: 191893.0000 - fn: 15.0000 - accuracy: 0.9821 - precision: 0.9654 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0785 - val_tp: 131.0000 - val_fp: 1513.0000 - val_tn: 83782.0000 - val_fn: 17.0000 - val_accuracy: 0.9821 - val_precision: 0.0797 - val_recall: 0.8851 - val_auc: 0.9598 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9947
Epoch 109/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1233 - tp: 199002.0000 - fp: 6939.0000 - tn: 192081.0000 - fn: 18.0000 - accuracy: 0.9825 - precision: 0.9663 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0766 - val_tp: 131.0000 - val_fp: 1502.0000 - val_tn: 83793.0000 - val_fn: 17.0000 - val_accuracy: 0.9822 - val_precision: 0.0802 - val_recall: 0.8851 - val_auc: 0.9628 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9960
Epoch 110/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1096 - tp: 199007.0000 - fp: 6751.0000 - tn: 192269.0000 - fn: 13.0000 - accuracy: 0.9830 - precision: 0.9672 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.1011 - val_tp: 134.0000 - val_fp: 1906.0000 - val_tn: 83389.0000 - val_fn: 14.0000 - val_accuracy: 0.9775 - val_precision: 0.0657 - val_recall: 0.9054 - val_auc: 0.9596 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9926
Epoch 111/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1038 - tp: 199007.0000 - fp: 6865.0000 - tn: 192155.0000 - fn: 13.0000 - accuracy: 0.9827 - precision: 0.9667 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9947 - val_loss: 0.0798 - val_tp: 130.0000 - val_fp: 1435.0000 - val_tn: 83860.0000 - val_fn: 18.0000 - val_accuracy: 0.9830 - val_precision: 0.0831 - val_recall: 0.8784 - val_auc: 0.9611 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 112/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1194 - tp: 199005.0000 - fp: 6825.0000 - tn: 192195.0000 - fn: 15.0000 - accuracy: 0.9828 - precision: 0.9668 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0782 - val_tp: 130.0000 - val_fp: 1407.0000 - val_tn: 83888.0000 - val_fn: 18.0000 - val_accuracy: 0.9833 - val_precision: 0.0846 - val_recall: 0.8784 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 113/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1230 - tp: 199000.0000 - fp: 6943.0000 - tn: 192077.0000 - fn: 20.0000 - accuracy: 0.9825 - precision: 0.9663 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9952 - val_loss: 0.0920 - val_tp: 131.0000 - val_fp: 1623.0000 - val_tn: 83672.0000 - val_fn: 17.0000 - val_accuracy: 0.9808 - val_precision: 0.0747 - val_recall: 0.8851 - val_auc: 0.9613 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9933
Epoch 114/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1333 - tp: 199002.0000 - fp: 7270.0000 - tn: 191750.0000 - fn: 18.0000 - accuracy: 0.9817 - precision: 0.9648 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9947 - val_loss: 0.0772 - val_tp: 131.0000 - val_fp: 1457.0000 - val_tn: 83838.0000 - val_fn: 17.0000 - val_accuracy: 0.9827 - val_precision: 0.0825 - val_recall: 0.8851 - val_auc: 0.9621 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9936
Epoch 115/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1150 - tp: 199004.0000 - fp: 7305.0000 - tn: 191715.0000 - fn: 16.0000 - accuracy: 0.9816 - precision: 0.9646 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0814 - val_tp: 130.0000 - val_fp: 1553.0000 - val_tn: 83742.0000 - val_fn: 18.0000 - val_accuracy: 0.9816 - val_precision: 0.0772 - val_recall: 0.8784 - val_auc: 0.9598 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9934
Epoch 116/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1216 - tp: 199005.0000 - fp: 7058.0000 - tn: 191962.0000 - fn: 15.0000 - accuracy: 0.9822 - precision: 0.9657 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9952 - val_loss: 0.0743 - val_tp: 130.0000 - val_fp: 1383.0000 - val_tn: 83912.0000 - val_fn: 18.0000 - val_accuracy: 0.9836 - val_precision: 0.0859 - val_recall: 0.8784 - val_auc: 0.9597 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9943
Epoch 117/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1275 - tp: 199006.0000 - fp: 7730.0000 - tn: 191290.0000 - fn: 14.0000 - accuracy: 0.9805 - precision: 0.9626 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9947 - val_loss: 0.0725 - val_tp: 130.0000 - val_fp: 1214.0000 - val_tn: 84081.0000 - val_fn: 18.0000 - val_accuracy: 0.9856 - val_precision: 0.0967 - val_recall: 0.8784 - val_auc: 0.9605 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9933
Epoch 118/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.1242 - tp: 199007.0000 - fp: 6435.0000 - tn: 192585.0000 - fn: 13.0000 - accuracy: 0.9838 - precision: 0.9687 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0764 - val_tp: 130.0000 - val_fp: 1292.0000 - val_tn: 84003.0000 - val_fn: 18.0000 - val_accuracy: 0.9847 - val_precision: 0.0914 - val_recall: 0.8784 - val_auc: 0.9603 - val_sensitivity: 0.0000e+00 - val_specificity: 0.99270 - fp: 873.0000 - tn: 30339.0000 - fn: 0.0000e+00 - accuracy: 0.9860 - 
Epoch 119/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1271 - tp: 199003.0000 - fp: 6690.0000 - tn: 192330.0000 - fn: 17.0000 - accuracy: 0.9831 - precision: 0.9675 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0848 - val_tp: 132.0000 - val_fp: 1608.0000 - val_tn: 83687.0000 - val_fn: 16.0000 - val_accuracy: 0.9810 - val_precision: 0.0759 - val_recall: 0.8919 - val_auc: 0.9590 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9944
Epoch 120/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1359 - tp: 199000.0000 - fp: 7775.0000 - tn: 191245.0000 - fn: 20.0000 - accuracy: 0.9804 - precision: 0.9624 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9943 - val_loss: 0.0910 - val_tp: 132.0000 - val_fp: 1925.0000 - val_tn: 83370.0000 - val_fn: 16.0000 - val_accuracy: 0.9773 - val_precision: 0.0642 - val_recall: 0.8919 - val_auc: 0.9640 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9961
Epoch 121/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1209 - tp: 199009.0000 - fp: 7841.0000 - tn: 191179.0000 - fn: 11.0000 - accuracy: 0.9803 - precision: 0.9621 - recall: 0.9999 - auc: 0.9972 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0877 - val_tp: 131.0000 - val_fp: 1684.0000 - val_tn: 83611.0000 - val_fn: 17.0000 - val_accuracy: 0.9801 - val_precision: 0.0722 - val_recall: 0.8851 - val_auc: 0.9615 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9935
Epoch 122/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1077 - tp: 199007.0000 - fp: 6988.0000 - tn: 192032.0000 - fn: 13.0000 - accuracy: 0.9824 - precision: 0.9661 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9948 - val_loss: 0.0784 - val_tp: 132.0000 - val_fp: 1338.0000 - val_tn: 83957.0000 - val_fn: 16.0000 - val_accuracy: 0.9842 - val_precision: 0.0898 - val_recall: 0.8919 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9928
Epoch 123/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1237 - tp: 199003.0000 - fp: 6575.0000 - tn: 192445.0000 - fn: 17.0000 - accuracy: 0.9834 - precision: 0.9680 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0816 - val_tp: 133.0000 - val_fp: 1533.0000 - val_tn: 83762.0000 - val_fn: 15.0000 - val_accuracy: 0.9819 - val_precision: 0.0798 - val_recall: 0.8986 - val_auc: 0.9618 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9961
Epoch 124/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1321 - tp: 199002.0000 - fp: 7835.0000 - tn: 191185.0000 - fn: 18.0000 - accuracy: 0.9803 - precision: 0.9621 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0848 - val_tp: 131.0000 - val_fp: 1650.0000 - val_tn: 83645.0000 - val_fn: 17.0000 - val_accuracy: 0.9805 - val_precision: 0.0736 - val_recall: 0.8851 - val_auc: 0.9650 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9959
Epoch 125/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1097 - tp: 199008.0000 - fp: 7033.0000 - tn: 191987.0000 - fn: 12.0000 - accuracy: 0.9823 - precision: 0.9659 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0832 - val_tp: 130.0000 - val_fp: 1480.0000 - val_tn: 83815.0000 - val_fn: 18.0000 - val_accuracy: 0.9825 - val_precision: 0.0807 - val_recall: 0.8784 - val_auc: 0.9632 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 126/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1168 - tp: 199003.0000 - fp: 7100.0000 - tn: 191920.0000 - fn: 17.0000 - accuracy: 0.9821 - precision: 0.9656 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0788 - val_tp: 131.0000 - val_fp: 1441.0000 - val_tn: 83854.0000 - val_fn: 17.0000 - val_accuracy: 0.9829 - val_precision: 0.0833 - val_recall: 0.8851 - val_auc: 0.9632 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9945
Epoch 127/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1102 - tp: 199004.0000 - fp: 6587.0000 - tn: 192433.0000 - fn: 16.0000 - accuracy: 0.9834 - precision: 0.9680 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0822 - val_tp: 130.0000 - val_fp: 1436.0000 - val_tn: 83859.0000 - val_fn: 18.0000 - val_accuracy: 0.9830 - val_precision: 0.0830 - val_recall: 0.8784 - val_auc: 0.9596 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 128/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1127 - tp: 199004.0000 - fp: 6701.0000 - tn: 192319.0000 - fn: 16.0000 - accuracy: 0.9831 - precision: 0.9674 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0942 - val_tp: 132.0000 - val_fp: 1886.0000 - val_tn: 83409.0000 - val_fn: 16.0000 - val_accuracy: 0.9777 - val_precision: 0.0654 - val_recall: 0.8919 - val_auc: 0.9593 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9941
Epoch 129/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1031 - tp: 199010.0000 - fp: 7398.0000 - tn: 191622.0000 - fn: 10.0000 - accuracy: 0.9814 - precision: 0.9642 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0792 - val_tp: 130.0000 - val_fp: 1341.0000 - val_tn: 83954.0000 - val_fn: 18.0000 - val_accuracy: 0.9841 - val_precision: 0.0884 - val_recall: 0.8784 - val_auc: 0.9609 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 130/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1150 - tp: 199006.0000 - fp: 6223.0000 - tn: 192797.0000 - fn: 14.0000 - accuracy: 0.9843 - precision: 0.9697 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0872 - val_tp: 131.0000 - val_fp: 1493.0000 - val_tn: 83802.0000 - val_fn: 17.0000 - val_accuracy: 0.9823 - val_precision: 0.0807 - val_recall: 0.8851 - val_auc: 0.9603 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9938
Epoch 131/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1248 - tp: 199005.0000 - fp: 6447.0000 - tn: 192573.0000 - fn: 15.0000 - accuracy: 0.9838 - precision: 0.9686 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0778 - val_tp: 131.0000 - val_fp: 1403.0000 - val_tn: 83892.0000 - val_fn: 17.0000 - val_accuracy: 0.9834 - val_precision: 0.0854 - val_recall: 0.8851 - val_auc: 0.9636 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9945
Epoch 132/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1224 - tp: 199004.0000 - fp: 6344.0000 - tn: 192676.0000 - fn: 16.0000 - accuracy: 0.9840 - precision: 0.9691 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9946 - val_loss: 0.0712 - val_tp: 131.0000 - val_fp: 1236.0000 - val_tn: 84059.0000 - val_fn: 17.0000 - val_accuracy: 0.9853 - val_precision: 0.0958 - val_recall: 0.8851 - val_auc: 0.9611 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9949
Epoch 133/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1134 - tp: 199007.0000 - fp: 6378.0000 - tn: 192642.0000 - fn: 13.0000 - accuracy: 0.9839 - precision: 0.9689 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0839 - val_tp: 130.0000 - val_fp: 1450.0000 - val_tn: 83845.0000 - val_fn: 18.0000 - val_accuracy: 0.9828 - val_precision: 0.0823 - val_recall: 0.8784 - val_auc: 0.9647 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 134/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1106 - tp: 199007.0000 - fp: 7225.0000 - tn: 191795.0000 - fn: 13.0000 - accuracy: 0.9818 - precision: 0.9650 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9953 - val_loss: 0.0817 - val_tp: 130.0000 - val_fp: 1392.0000 - val_tn: 83903.0000 - val_fn: 18.0000 - val_accuracy: 0.9835 - val_precision: 0.0854 - val_recall: 0.8784 - val_auc: 0.9595 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 135/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1144 - tp: 199006.0000 - fp: 6406.0000 - tn: 192614.0000 - fn: 14.0000 - accuracy: 0.9839 - precision: 0.9688 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0764 - val_tp: 130.0000 - val_fp: 1277.0000 - val_tn: 84018.0000 - val_fn: 18.0000 - val_accuracy: 0.9848 - val_precision: 0.0924 - val_recall: 0.8784 - val_auc: 0.9664 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9945
Epoch 136/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1137 - tp: 199008.0000 - fp: 6782.0000 - tn: 192238.0000 - fn: 12.0000 - accuracy: 0.9829 - precision: 0.9670 - recall: 0.9999 - auc: 0.9969 - sensitivity: 0.0000e+00 - specificity: 0.9939 - val_loss: 0.0786 - val_tp: 130.0000 - val_fp: 1399.0000 - val_tn: 83896.0000 - val_fn: 18.0000 - val_accuracy: 0.9834 - val_precision: 0.0850 - val_recall: 0.8784 - val_auc: 0.9603 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9950
Epoch 137/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1039 - tp: 199006.0000 - fp: 6049.0000 - tn: 192971.0000 - fn: 14.0000 - accuracy: 0.9848 - precision: 0.9705 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0832 - val_tp: 131.0000 - val_fp: 1535.0000 - val_tn: 83760.0000 - val_fn: 17.0000 - val_accuracy: 0.9818 - val_precision: 0.0786 - val_recall: 0.8851 - val_auc: 0.9598 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9954
Epoch 138/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1127 - tp: 199006.0000 - fp: 7335.0000 - tn: 191685.0000 - fn: 14.0000 - accuracy: 0.9815 - precision: 0.9645 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0726 - val_tp: 130.0000 - val_fp: 1190.0000 - val_tn: 84105.0000 - val_fn: 18.0000 - val_accuracy: 0.9859 - val_precision: 0.0985 - val_recall: 0.8784 - val_auc: 0.9620 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9948
Epoch 139/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1125 - tp: 199007.0000 - fp: 6535.0000 - tn: 192485.0000 - fn: 13.0000 - accuracy: 0.9835 - precision: 0.9682 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0784 - val_tp: 130.0000 - val_fp: 1286.0000 - val_tn: 84009.0000 - val_fn: 18.0000 - val_accuracy: 0.9847 - val_precision: 0.0918 - val_recall: 0.8784 - val_auc: 0.9624 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9953
Epoch 140/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.1200 - tp: 199006.0000 - fp: 7036.0000 - tn: 191984.0000 - fn: 14.0000 - accuracy: 0.9823 - precision: 0.9659 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0896 - val_tp: 132.0000 - val_fp: 1521.0000 - val_tn: 83774.0000 - val_fn: 16.0000 - val_accuracy: 0.9820 - val_precision: 0.0799 - val_recall: 0.8919 - val_auc: 0.9634 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9958
Epoch 141/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1137 - tp: 199008.0000 - fp: 7130.0000 - tn: 191890.0000 - fn: 12.0000 - accuracy: 0.9821 - precision: 0.9654 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0805 - val_tp: 129.0000 - val_fp: 1320.0000 - val_tn: 83975.0000 - val_fn: 19.0000 - val_accuracy: 0.9843 - val_precision: 0.0890 - val_recall: 0.8716 - val_auc: 0.9613 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 142/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1086 - tp: 199004.0000 - fp: 6328.0000 - tn: 192692.0000 - fn: 16.0000 - accuracy: 0.9841 - precision: 0.9692 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0756 - val_tp: 131.0000 - val_fp: 1295.0000 - val_tn: 84000.0000 - val_fn: 17.0000 - val_accuracy: 0.9846 - val_precision: 0.0919 - val_recall: 0.8851 - val_auc: 0.9620 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9960
Epoch 143/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1186 - tp: 199006.0000 - fp: 6548.0000 - tn: 192472.0000 - fn: 14.0000 - accuracy: 0.9835 - precision: 0.9681 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0726 - val_tp: 130.0000 - val_fp: 1198.0000 - val_tn: 84097.0000 - val_fn: 18.0000 - val_accuracy: 0.9858 - val_precision: 0.0979 - val_recall: 0.8784 - val_auc: 0.9596 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9945
Epoch 144/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.1117 - tp: 199003.0000 - fp: 6353.0000 - tn: 192667.0000 - fn: 17.0000 - accuracy: 0.9840 - precision: 0.9691 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0869 - val_tp: 132.0000 - val_fp: 1572.0000 - val_tn: 83723.0000 - val_fn: 16.0000 - val_accuracy: 0.9814 - val_precision: 0.0775 - val_recall: 0.8919 - val_auc: 0.9646 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9941
Epoch 145/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1055 - tp: 199008.0000 - fp: 6833.0000 - tn: 192187.0000 - fn: 12.0000 - accuracy: 0.9828 - precision: 0.9668 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0884 - val_tp: 131.0000 - val_fp: 1389.0000 - val_tn: 83906.0000 - val_fn: 17.0000 - val_accuracy: 0.9835 - val_precision: 0.0862 - val_recall: 0.8851 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9929
Epoch 146/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.1102 - tp: 199006.0000 - fp: 6257.0000 - tn: 192763.0000 - fn: 14.0000 - accuracy: 0.9842 - precision: 0.9695 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0692 - val_tp: 129.0000 - val_fp: 1181.0000 - val_tn: 84114.0000 - val_fn: 19.0000 - val_accuracy: 0.9860 - val_precision: 0.0985 - val_recall: 0.8716 - val_auc: 0.9601 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9954
Epoch 147/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1176 - tp: 199004.0000 - fp: 6214.0000 - tn: 192806.0000 - fn: 16.0000 - accuracy: 0.9843 - precision: 0.9697 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0762 - val_tp: 129.0000 - val_fp: 1182.0000 - val_tn: 84113.0000 - val_fn: 19.0000 - val_accuracy: 0.9859 - val_precision: 0.0984 - val_recall: 0.8716 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9960
Epoch 148/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0951 - tp: 199007.0000 - fp: 5628.0000 - tn: 193392.0000 - fn: 13.0000 - accuracy: 0.9858 - precision: 0.9725 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0660 - val_tp: 130.0000 - val_fp: 1081.0000 - val_tn: 84214.0000 - val_fn: 18.0000 - val_accuracy: 0.9871 - val_precision: 0.1073 - val_recall: 0.8784 - val_auc: 0.9607 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9960
Epoch 149/200
398040/398040 [==============================] - 5s 12us/sample - loss: 0.1057 - tp: 199007.0000 - fp: 5912.0000 - tn: 193108.0000 - fn: 13.0000 - accuracy: 0.9851 - precision: 0.9711 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9953 - val_loss: 0.0710 - val_tp: 130.0000 - val_fp: 1165.0000 - val_tn: 84130.0000 - val_fn: 18.0000 - val_accuracy: 0.9862 - val_precision: 0.1004 - val_recall: 0.8784 - val_auc: 0.9595 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9948
Epoch 150/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1145 - tp: 199006.0000 - fp: 6062.0000 - tn: 192958.0000 - fn: 14.0000 - accuracy: 0.9847 - precision: 0.9704 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0678 - val_tp: 130.0000 - val_fp: 1087.0000 - val_tn: 84208.0000 - val_fn: 18.0000 - val_accuracy: 0.9871 - val_precision: 0.1068 - val_recall: 0.8784 - val_auc: 0.9619 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9954
Epoch 151/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1067 - tp: 199004.0000 - fp: 6723.0000 - tn: 192297.0000 - fn: 16.0000 - accuracy: 0.9831 - precision: 0.9673 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0742 - val_tp: 130.0000 - val_fp: 1200.0000 - val_tn: 84095.0000 - val_fn: 18.0000 - val_accuracy: 0.9857 - val_precision: 0.0977 - val_recall: 0.8784 - val_auc: 0.9593 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9958
Epoch 152/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1070 - tp: 199006.0000 - fp: 5831.0000 - tn: 193189.0000 - fn: 14.0000 - accuracy: 0.9853 - precision: 0.9715 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0904 - val_tp: 131.0000 - val_fp: 1400.0000 - val_tn: 83895.0000 - val_fn: 17.0000 - val_accuracy: 0.9834 - val_precision: 0.0856 - val_recall: 0.8851 - val_auc: 0.9586 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 153/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1119 - tp: 199007.0000 - fp: 6884.0000 - tn: 192136.0000 - fn: 13.0000 - accuracy: 0.9827 - precision: 0.9666 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0707 - val_tp: 130.0000 - val_fp: 1184.0000 - val_tn: 84111.0000 - val_fn: 18.0000 - val_accuracy: 0.9859 - val_precision: 0.0989 - val_recall: 0.8784 - val_auc: 0.9624 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9958
Epoch 154/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1064 - tp: 199007.0000 - fp: 6773.0000 - tn: 192247.0000 - fn: 13.0000 - accuracy: 0.9830 - precision: 0.9671 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9948 - val_loss: 0.0878 - val_tp: 131.0000 - val_fp: 1495.0000 - val_tn: 83800.0000 - val_fn: 17.0000 - val_accuracy: 0.9823 - val_precision: 0.0806 - val_recall: 0.8851 - val_auc: 0.9601 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9922
Epoch 155/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1189 - tp: 199004.0000 - fp: 6583.0000 - tn: 192437.0000 - fn: 16.0000 - accuracy: 0.9834 - precision: 0.9680 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0795 - val_tp: 132.0000 - val_fp: 1519.0000 - val_tn: 83776.0000 - val_fn: 16.0000 - val_accuracy: 0.9820 - val_precision: 0.0800 - val_recall: 0.8919 - val_auc: 0.9659 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9956
Epoch 156/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1114 - tp: 199006.0000 - fp: 6815.0000 - tn: 192205.0000 - fn: 14.0000 - accuracy: 0.9828 - precision: 0.9669 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9958 - val_loss: 0.0750 - val_tp: 131.0000 - val_fp: 1290.0000 - val_tn: 84005.0000 - val_fn: 17.0000 - val_accuracy: 0.9847 - val_precision: 0.0922 - val_recall: 0.8851 - val_auc: 0.9663 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9950
Epoch 157/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0954 - tp: 199008.0000 - fp: 5768.0000 - tn: 193252.0000 - fn: 12.0000 - accuracy: 0.9855 - precision: 0.9718 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0712 - val_tp: 130.0000 - val_fp: 1173.0000 - val_tn: 84122.0000 - val_fn: 18.0000 - val_accuracy: 0.9861 - val_precision: 0.0998 - val_recall: 0.8784 - val_auc: 0.9603 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9959
Epoch 158/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1097 - tp: 199005.0000 - fp: 5853.0000 - tn: 193167.0000 - fn: 15.0000 - accuracy: 0.9853 - precision: 0.9714 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9958 - val_loss: 0.0743 - val_tp: 130.0000 - val_fp: 1257.0000 - val_tn: 84038.0000 - val_fn: 18.0000 - val_accuracy: 0.9851 - val_precision: 0.0937 - val_recall: 0.8784 - val_auc: 0.9646 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9957
Epoch 159/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1197 - tp: 199004.0000 - fp: 6408.0000 - tn: 192612.0000 - fn: 16.0000 - accuracy: 0.9839 - precision: 0.9688 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9953 - val_loss: 0.0795 - val_tp: 130.0000 - val_fp: 1359.0000 - val_tn: 83936.0000 - val_fn: 18.0000 - val_accuracy: 0.9839 - val_precision: 0.0873 - val_recall: 0.8784 - val_auc: 0.9639 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 160/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1071 - tp: 199005.0000 - fp: 6310.0000 - tn: 192710.0000 - fn: 15.0000 - accuracy: 0.9841 - precision: 0.9693 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0810 - val_tp: 130.0000 - val_fp: 1441.0000 - val_tn: 83854.0000 - val_fn: 18.0000 - val_accuracy: 0.9829 - val_precision: 0.0827 - val_recall: 0.8784 - val_auc: 0.9586 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9951
Epoch 161/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0988 - tp: 199008.0000 - fp: 5703.0000 - tn: 193317.0000 - fn: 12.0000 - accuracy: 0.9856 - precision: 0.9721 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9957 - val_loss: 0.0747 - val_tp: 130.0000 - val_fp: 1152.0000 - val_tn: 84143.0000 - val_fn: 18.0000 - val_accuracy: 0.9863 - val_precision: 0.1014 - val_recall: 0.8784 - val_auc: 0.9601 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9928
Epoch 162/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1113 - tp: 199005.0000 - fp: 6664.0000 - tn: 192356.0000 - fn: 15.0000 - accuracy: 0.9832 - precision: 0.9676 - recall: 0.9999 - auc: 0.9973 - sensitivity: 0.0000e+00 - specificity: 0.9945 - val_loss: 0.0745 - val_tp: 130.0000 - val_fp: 1267.0000 - val_tn: 84028.0000 - val_fn: 18.0000 - val_accuracy: 0.9850 - val_precision: 0.0931 - val_recall: 0.8784 - val_auc: 0.9596 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9935
Epoch 163/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1091 - tp: 199006.0000 - fp: 5854.0000 - tn: 193166.0000 - fn: 14.0000 - accuracy: 0.9853 - precision: 0.9714 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0672 - val_tp: 130.0000 - val_fp: 1161.0000 - val_tn: 84134.0000 - val_fn: 18.0000 - val_accuracy: 0.9862 - val_precision: 0.1007 - val_recall: 0.8784 - val_auc: 0.9628 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9964
Epoch 164/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1012 - tp: 199005.0000 - fp: 6382.0000 - tn: 192638.0000 - fn: 15.0000 - accuracy: 0.9839 - precision: 0.9689 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0729 - val_tp: 130.0000 - val_fp: 1222.0000 - val_tn: 84073.0000 - val_fn: 18.0000 - val_accuracy: 0.9855 - val_precision: 0.0962 - val_recall: 0.8784 - val_auc: 0.9602 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9941
Epoch 165/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0991 - tp: 199007.0000 - fp: 5742.0000 - tn: 193278.0000 - fn: 13.0000 - accuracy: 0.9855 - precision: 0.9720 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0853 - val_tp: 130.0000 - val_fp: 1346.0000 - val_tn: 83949.0000 - val_fn: 18.0000 - val_accuracy: 0.9840 - val_precision: 0.0881 - val_recall: 0.8784 - val_auc: 0.9584 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9926
Epoch 166/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0985 - tp: 199008.0000 - fp: 5732.0000 - tn: 193288.0000 - fn: 12.0000 - accuracy: 0.9856 - precision: 0.9720 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0764 - val_tp: 130.0000 - val_fp: 1266.0000 - val_tn: 84029.0000 - val_fn: 18.0000 - val_accuracy: 0.9850 - val_precision: 0.0931 - val_recall: 0.8784 - val_auc: 0.9639 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9932
Epoch 167/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0973 - tp: 199009.0000 - fp: 5338.0000 - tn: 193682.0000 - fn: 11.0000 - accuracy: 0.9866 - precision: 0.9739 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0724 - val_tp: 130.0000 - val_fp: 1194.0000 - val_tn: 84101.0000 - val_fn: 18.0000 - val_accuracy: 0.9858 - val_precision: 0.0982 - val_recall: 0.8784 - val_auc: 0.9644 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 168/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1072 - tp: 199007.0000 - fp: 6061.0000 - tn: 192959.0000 - fn: 13.0000 - accuracy: 0.9847 - precision: 0.9704 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9957 - val_loss: 0.0669 - val_tp: 130.0000 - val_fp: 1179.0000 - val_tn: 84116.0000 - val_fn: 18.0000 - val_accuracy: 0.9860 - val_precision: 0.0993 - val_recall: 0.8784 - val_auc: 0.9656 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9959
Epoch 169/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1096 - tp: 199005.0000 - fp: 6329.0000 - tn: 192691.0000 - fn: 15.0000 - accuracy: 0.9841 - precision: 0.9692 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0813 - val_tp: 130.0000 - val_fp: 1455.0000 - val_tn: 83840.0000 - val_fn: 18.0000 - val_accuracy: 0.9828 - val_precision: 0.0820 - val_recall: 0.8784 - val_auc: 0.9647 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9939
Epoch 170/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1046 - tp: 199006.0000 - fp: 6827.0000 - tn: 192193.0000 - fn: 14.0000 - accuracy: 0.9828 - precision: 0.9668 - recall: 0.9999 - auc: 0.9974 - sensitivity: 0.0000e+00 - specificity: 0.9949 - val_loss: 0.0799 - val_tp: 131.0000 - val_fp: 1412.0000 - val_tn: 83883.0000 - val_fn: 17.0000 - val_accuracy: 0.9833 - val_precision: 0.0849 - val_recall: 0.8851 - val_auc: 0.9649 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 171/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1126 - tp: 199003.0000 - fp: 6598.0000 - tn: 192422.0000 - fn: 17.0000 - accuracy: 0.9834 - precision: 0.9679 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9953 - val_loss: 0.0746 - val_tp: 131.0000 - val_fp: 1446.0000 - val_tn: 83849.0000 - val_fn: 17.0000 - val_accuracy: 0.9829 - val_precision: 0.0831 - val_recall: 0.8851 - val_auc: 0.9615 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9960
Epoch 172/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1052 - tp: 199008.0000 - fp: 6343.0000 - tn: 192677.0000 - fn: 12.0000 - accuracy: 0.9840 - precision: 0.9691 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0753 - val_tp: 130.0000 - val_fp: 1406.0000 - val_tn: 83889.0000 - val_fn: 18.0000 - val_accuracy: 0.9833 - val_precision: 0.0846 - val_recall: 0.8784 - val_auc: 0.9618 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9937
Epoch 173/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1081 - tp: 199005.0000 - fp: 6007.0000 - tn: 193013.0000 - fn: 15.0000 - accuracy: 0.9849 - precision: 0.9707 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0810 - val_tp: 130.0000 - val_fp: 1569.0000 - val_tn: 83726.0000 - val_fn: 18.0000 - val_accuracy: 0.9814 - val_precision: 0.0765 - val_recall: 0.8784 - val_auc: 0.9647 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9938
Epoch 174/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1160 - tp: 199002.0000 - fp: 7020.0000 - tn: 192000.0000 - fn: 18.0000 - accuracy: 0.9823 - precision: 0.9659 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9953 - val_loss: 0.0842 - val_tp: 132.0000 - val_fp: 1669.0000 - val_tn: 83626.0000 - val_fn: 16.0000 - val_accuracy: 0.9803 - val_precision: 0.0733 - val_recall: 0.8919 - val_auc: 0.9676 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 175/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1007 - tp: 199010.0000 - fp: 7032.0000 - tn: 191988.0000 - fn: 10.0000 - accuracy: 0.9823 - precision: 0.9659 - recall: 0.9999 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9953 - val_loss: 0.0681 - val_tp: 130.0000 - val_fp: 1209.0000 - val_tn: 84086.0000 - val_fn: 18.0000 - val_accuracy: 0.9856 - val_precision: 0.0971 - val_recall: 0.8784 - val_auc: 0.9663 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 176/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1106 - tp: 199007.0000 - fp: 6206.0000 - tn: 192814.0000 - fn: 13.0000 - accuracy: 0.9844 - precision: 0.9698 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0664 - val_tp: 131.0000 - val_fp: 1157.0000 - val_tn: 84138.0000 - val_fn: 17.0000 - val_accuracy: 0.9863 - val_precision: 0.1017 - val_recall: 0.8851 - val_auc: 0.9626 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9966
Epoch 177/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1023 - tp: 199008.0000 - fp: 5996.0000 - tn: 193024.0000 - fn: 12.0000 - accuracy: 0.9849 - precision: 0.9708 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0721 - val_tp: 131.0000 - val_fp: 1273.0000 - val_tn: 84022.0000 - val_fn: 17.0000 - val_accuracy: 0.9849 - val_precision: 0.0933 - val_recall: 0.8851 - val_auc: 0.9656 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 178/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1040 - tp: 199007.0000 - fp: 5857.0000 - tn: 193163.0000 - fn: 13.0000 - accuracy: 0.9853 - precision: 0.9714 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0690 - val_tp: 131.0000 - val_fp: 1206.0000 - val_tn: 84089.0000 - val_fn: 17.0000 - val_accuracy: 0.9857 - val_precision: 0.0980 - val_recall: 0.8851 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9967
Epoch 179/200
398040/398040 [==============================] - 5s 11us/sample - loss: 0.1123 - tp: 199007.0000 - fp: 6952.0000 - tn: 192068.0000 - fn: 13.0000 - accuracy: 0.9825 - precision: 0.9662 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0685 - val_tp: 130.0000 - val_fp: 1126.0000 - val_tn: 84169.0000 - val_fn: 18.0000 - val_accuracy: 0.9866 - val_precision: 0.1035 - val_recall: 0.8784 - val_auc: 0.9650 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 180/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1040 - tp: 199010.0000 - fp: 6039.0000 - tn: 192981.0000 - fn: 10.0000 - accuracy: 0.9848 - precision: 0.9705 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9954 - val_loss: 0.0631 - val_tp: 130.0000 - val_fp: 1003.0000 - val_tn: 84292.0000 - val_fn: 18.0000 - val_accuracy: 0.9881 - val_precision: 0.1147 - val_recall: 0.8784 - val_auc: 0.9606 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9965
Epoch 181/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1180 - tp: 199004.0000 - fp: 5868.0000 - tn: 193152.0000 - fn: 16.0000 - accuracy: 0.9852 - precision: 0.9714 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0631 - val_tp: 130.0000 - val_fp: 1035.0000 - val_tn: 84260.0000 - val_fn: 18.0000 - val_accuracy: 0.9877 - val_precision: 0.1116 - val_recall: 0.8784 - val_auc: 0.9631 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9968
Epoch 182/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0952 - tp: 199010.0000 - fp: 5593.0000 - tn: 193427.0000 - fn: 10.0000 - accuracy: 0.9859 - precision: 0.9727 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9957 - val_loss: 0.0576 - val_tp: 129.0000 - val_fp: 915.0000 - val_tn: 84380.0000 - val_fn: 19.0000 - val_accuracy: 0.9891 - val_precision: 0.1236 - val_recall: 0.8716 - val_auc: 0.9612 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9965
Epoch 183/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1066 - tp: 199007.0000 - fp: 5487.0000 - tn: 193533.0000 - fn: 13.0000 - accuracy: 0.9862 - precision: 0.9732 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9959 - val_loss: 0.0637 - val_tp: 130.0000 - val_fp: 1015.0000 - val_tn: 84280.0000 - val_fn: 18.0000 - val_accuracy: 0.9879 - val_precision: 0.1135 - val_recall: 0.8784 - val_auc: 0.9653 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9966
Epoch 184/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1123 - tp: 199003.0000 - fp: 6086.0000 - tn: 192934.0000 - fn: 17.0000 - accuracy: 0.9847 - precision: 0.9703 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9959 - val_loss: 0.0772 - val_tp: 132.0000 - val_fp: 1280.0000 - val_tn: 84015.0000 - val_fn: 16.0000 - val_accuracy: 0.9848 - val_precision: 0.0935 - val_recall: 0.8919 - val_auc: 0.9670 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9931
Epoch 185/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1016 - tp: 199005.0000 - fp: 6438.0000 - tn: 192582.0000 - fn: 15.0000 - accuracy: 0.9838 - precision: 0.9687 - recall: 0.9999 - auc: 0.9977 - sensitivity: 0.0000e+00 - specificity: 0.9955 - val_loss: 0.0708 - val_tp: 131.0000 - val_fp: 1114.0000 - val_tn: 84181.0000 - val_fn: 17.0000 - val_accuracy: 0.9868 - val_precision: 0.1052 - val_recall: 0.8851 - val_auc: 0.9628 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9938
Epoch 186/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0960 - tp: 199011.0000 - fp: 5035.0000 - tn: 193985.0000 - fn: 9.0000 - accuracy: 0.9873 - precision: 0.9753 - recall: 1.0000 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9958 - val_loss: 0.0681 - val_tp: 131.0000 - val_fp: 1109.0000 - val_tn: 84186.0000 - val_fn: 17.0000 - val_accuracy: 0.9868 - val_precision: 0.1056 - val_recall: 0.8851 - val_auc: 0.9687 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9942
Epoch 187/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0905 - tp: 199005.0000 - fp: 5555.0000 - tn: 193465.0000 - fn: 15.0000 - accuracy: 0.9860 - precision: 0.9728 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9959 - val_loss: 0.0755 - val_tp: 131.0000 - val_fp: 1317.0000 - val_tn: 83978.0000 - val_fn: 17.0000 - val_accuracy: 0.9844 - val_precision: 0.0905 - val_recall: 0.8851 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 188/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1064 - tp: 199006.0000 - fp: 6059.0000 - tn: 192961.0000 - fn: 14.0000 - accuracy: 0.9847 - precision: 0.9705 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9957 - val_loss: 0.0648 - val_tp: 130.0000 - val_fp: 1166.0000 - val_tn: 84129.0000 - val_fn: 18.0000 - val_accuracy: 0.9861 - val_precision: 0.1003 - val_recall: 0.8784 - val_auc: 0.9631 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9948
Epoch 189/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1022 - tp: 199010.0000 - fp: 6082.0000 - tn: 192938.0000 - fn: 10.0000 - accuracy: 0.9847 - precision: 0.9703 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0779 - val_tp: 130.0000 - val_fp: 1331.0000 - val_tn: 83964.0000 - val_fn: 18.0000 - val_accuracy: 0.9842 - val_precision: 0.0890 - val_recall: 0.8784 - val_auc: 0.9603 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9932
Epoch 190/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1038 - tp: 199006.0000 - fp: 6062.0000 - tn: 192958.0000 - fn: 14.0000 - accuracy: 0.9847 - precision: 0.9704 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0591 - val_tp: 129.0000 - val_fp: 994.0000 - val_tn: 84301.0000 - val_fn: 19.0000 - val_accuracy: 0.9881 - val_precision: 0.1149 - val_recall: 0.8716 - val_auc: 0.9635 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9966
Epoch 191/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1165 - tp: 199004.0000 - fp: 5991.0000 - tn: 193029.0000 - fn: 16.0000 - accuracy: 0.9849 - precision: 0.9708 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0846 - val_tp: 129.0000 - val_fp: 1559.0000 - val_tn: 83736.0000 - val_fn: 19.0000 - val_accuracy: 0.9815 - val_precision: 0.0764 - val_recall: 0.8716 - val_auc: 0.9669 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9934
Epoch 192/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1151 - tp: 199001.0000 - fp: 6215.0000 - tn: 192805.0000 - fn: 19.0000 - accuracy: 0.9843 - precision: 0.9697 - recall: 0.9999 - auc: 0.9980 - sensitivity: 0.0000e+00 - specificity: 0.9960 - val_loss: 0.0692 - val_tp: 131.0000 - val_fp: 1344.0000 - val_tn: 83951.0000 - val_fn: 17.0000 - val_accuracy: 0.9841 - val_precision: 0.0888 - val_recall: 0.8851 - val_auc: 0.9664 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9968
Epoch 193/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1267 - tp: 199000.0000 - fp: 7322.0000 - tn: 191698.0000 - fn: 20.0000 - accuracy: 0.9816 - precision: 0.9645 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0735 - val_tp: 131.0000 - val_fp: 1393.0000 - val_tn: 83902.0000 - val_fn: 17.0000 - val_accuracy: 0.9835 - val_precision: 0.0860 - val_recall: 0.8851 - val_auc: 0.9654 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9964
Epoch 194/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1005 - tp: 199008.0000 - fp: 6187.0000 - tn: 192833.0000 - fn: 12.0000 - accuracy: 0.9844 - precision: 0.9698 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0655 - val_tp: 130.0000 - val_fp: 1169.0000 - val_tn: 84126.0000 - val_fn: 18.0000 - val_accuracy: 0.9861 - val_precision: 0.1001 - val_recall: 0.8784 - val_auc: 0.9622 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9962
Epoch 195/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1165 - tp: 199007.0000 - fp: 6944.0000 - tn: 192076.0000 - fn: 13.0000 - accuracy: 0.9825 - precision: 0.9663 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9950 - val_loss: 0.0697 - val_tp: 130.0000 - val_fp: 1148.0000 - val_tn: 84147.0000 - val_fn: 18.0000 - val_accuracy: 0.9864 - val_precision: 0.1017 - val_recall: 0.8784 - val_auc: 0.9625 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9940
Epoch 196/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0931 - tp: 199007.0000 - fp: 5635.0000 - tn: 193385.0000 - fn: 13.0000 - accuracy: 0.9858 - precision: 0.9725 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9957 - val_loss: 0.0572 - val_tp: 129.0000 - val_fp: 993.0000 - val_tn: 84302.0000 - val_fn: 19.0000 - val_accuracy: 0.9882 - val_precision: 0.1150 - val_recall: 0.8716 - val_auc: 0.9610 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9967
Epoch 197/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.1090 - tp: 199005.0000 - fp: 5969.0000 - tn: 193051.0000 - fn: 15.0000 - accuracy: 0.9850 - precision: 0.9709 - recall: 0.9999 - auc: 0.9978 - sensitivity: 0.0000e+00 - specificity: 0.9956 - val_loss: 0.0666 - val_tp: 130.0000 - val_fp: 1145.0000 - val_tn: 84150.0000 - val_fn: 18.0000 - val_accuracy: 0.9864 - val_precision: 0.1020 - val_recall: 0.8784 - val_auc: 0.9598 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9961
Epoch 198/200
398040/398040 [==============================] - 4s 10us/sample - loss: 0.0964 - tp: 199010.0000 - fp: 5878.0000 - tn: 193142.0000 - fn: 10.0000 - accuracy: 0.9852 - precision: 0.9713 - recall: 0.9999 - auc: 0.9979 - sensitivity: 0.0000e+00 - specificity: 0.9958 - val_loss: 0.0666 - val_tp: 130.0000 - val_fp: 1111.0000 - val_tn: 84184.0000 - val_fn: 18.0000 - val_accuracy: 0.9868 - val_precision: 0.1048 - val_recall: 0.8784 - val_auc: 0.9600 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9958
Epoch 199/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.1075 - tp: 199005.0000 - fp: 5730.0000 - tn: 193290.0000 - fn: 15.0000 - accuracy: 0.9856 - precision: 0.9720 - recall: 0.9999 - auc: 0.9975 - sensitivity: 0.0000e+00 - specificity: 0.9951 - val_loss: 0.0692 - val_tp: 130.0000 - val_fp: 1285.0000 - val_tn: 84010.0000 - val_fn: 18.0000 - val_accuracy: 0.9848 - val_precision: 0.0919 - val_recall: 0.8784 - val_auc: 0.9623 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9947
Epoch 200/200
398040/398040 [==============================] - 4s 11us/sample - loss: 0.0883 - tp: 199013.0000 - fp: 5937.0000 - tn: 193083.0000 - fn: 7.0000 - accuracy: 0.9851 - precision: 0.9710 - recall: 1.0000 - auc: 0.9976 - sensitivity: 0.0000e+00 - specificity: 0.9952 - val_loss: 0.0733 - val_tp: 130.0000 - val_fp: 1198.0000 - val_tn: 84097.0000 - val_fn: 18.0000 - val_accuracy: 0.9858 - val_precision: 0.0979 - val_recall: 0.8784 - val_auc: 0.9611 - val_sensitivity: 0.0000e+00 - val_specificity: 0.9931cision: 0.9705 - recall: 1.0000 - auc: 0.9977 - sen

run_data = []
run_data.append(run_no_weight_history)
run_data.append(run_5_weight_history)
run_data.append(run_500_weight_history)

import matplotlib.pyplot as plt
def plot_data(run_data,train_param,test_param,xstep,yrangestart,yrangeend,ystep,title,ylabel,xlabel,legend,loc='upper right'):
    plt.figure(figsize=(10,10),facecolor='w')
    colorMap = ['red','blue','green']
    colorCounter = 0
    for hist_data in run_data: 
        plt.plot(hist_data.history[train_param],color=colorMap[colorCounter])
        plt.plot(hist_data.history[test_param],color=colorMap[colorCounter], linestyle="dotted")
        colorCounter = colorCounter + 1
    plt.xticks(np.arange(0, EPOCHS+1,step=xstep))
    plt.yticks(np.arange(yrangestart, yrangeend, step=ystep))
    plt.title(title)
    plt.ylabel(ylabel)
    plt.xlabel(xlabel)
    plt.legend(legend,loc=loc,prop={'size': 11})
    plt.show()

Training & Validation Loss

plot_data(run_data,'loss','val_loss',20,0,0.5,0.25,"Model's Training & Validation loss across epochs",'Loss','Epochs',['Train_No_Weight','Validation_No_Weight','Train_5_Weight','Validation_5_Weight','Train_100_Weight','Validation_100_Weight'],loc='upper right')

Training & Validation Accuracy

plot_data(run_data,'accuracy','val_accuracy',20,0.8, 1,0.1,"Model's Training & Validation accuracy across epochs",'Accuracy','Epochs',['Train_No_Weight','Validation_No_Weight','Train_5_Weight','Validation_5_Weight','Train_500_Weight','Validation_500_Weight'],loc='lower right')

Training & Validation Precision

plot_data(run_data,'precision','val_precision',20,0,1,0.1,"Model's Training & Validation Precision across epochs",'Precision','Epochs',['Train_No_Weight','Validation_No_Weight','Train_5_Weight','Validation_5_Weight','Train_500_Weight','Validation_500_Weight'],loc='lower middle')

Training & Validation Recall (Sensitivity)

plot_data(run_data,'recall','val_recall',20,0.75,1,0.01,"Model's Training & Validation recall (sensitivity) across epochs",'Recall (Sensitivity)','Epochs',['Train_No_Weight','Validation_No_Weight','Train_5_Weight','Validation_5_Weight','Train_500_Weight','Validation_500_Weight'],loc='upper right')

Training & Validation Specificity

plot_data(run_data,'specificity','val_specificity',20,0.97,1,0.005,"Model's Training & Validation Specificity across epochs",'Specificity','Epochs',['Train_No_Weight','Validation_No_Weight','Train_5_Weight','Validation_5_Weight','Train_100_Weight','Validation_100_Weight'],loc='lower right')

from sklearn import metrics
from sklearn.metrics import accuracy_score
from sklearn.metrics import precision_score
from sklearn.metrics import recall_score
from sklearn.metrics import f1_score
from sklearn.metrics import roc_auc_score
from sklearn.metrics import roc_curve
from sklearn.metrics import confusion_matrix
from sklearn.metrics import fbeta_score
from sklearn.metrics import auc
from sklearn.metrics import precision_recall_curve
import matplotlib.pyplot as plt

def calculate_and_plot_neural_network(model,color_index,title,label,algorithm):
    # predict probabilities for test set
    yhat_probs = model.predict(testX, verbose=0)
    # predict crisp classes for test set
    yhat_classes = model.predict_classes(testX, verbose=0)
    # reduce to 1d array
    yhat_probs = yhat_probs[:, 0]
    yhat_classes = yhat_classes[:, 0]

    fig, (ax1, ax2) = plt.subplots(1, 2)
    fig.set_size_inches(15, 6)    
    
    calculate_and_print_scores(testy, yhat_classes,yhat_probs)

    plot_roc_auc(ax1,testy, yhat_probs,label)
    
    plot_cm(ax2,testy, yhat_classes,color_index,title)
       
    create_barchart_label(label,algorithm)
    
    create_table_label(label,algorithm)

def plot_roc_auc_neural_network(ax, testy, yhat_probs,label):
    #Show ROC-AUC Plot
    fig = plt.figure(figsize=(5,5))
    # plot no skill roc curve
    plt.plot([0, 1], [0, 1], linestyle='--', label='No Skill')
    # calculate roc curve for model
    fpr, tpr, _ = roc_curve(testy, yhat_probs)
    # plot model roc curve
    plt.plot(fpr, tpr, marker='.', label = label)
    # axis labels
    plt.xlabel('False Positive Rate')
    plt.xlim(0,1)
    plt.ylabel('True Positive Rate')
    plt.ylim(0,1)
    plt.xticks(np.arange(0, 1, step=0.1))
    plt.yticks(np.arange(0, 1, step=0.1))
    #fig.suptitle('Receiver Operating Characteristics (ROC) Curve', fontsize=20)
    # show the legend
    plt.legend()
    # show the plot    

Results for model with no weight

calculate_and_plot_neural_network(model_no_weight,0,'Confusion Matrix for No Weight','Class Weight 1','Neural Network')

Accuracy: 0.999134
Precision: 0.717647
Sensitivity AKA Recall: 0.824324
F1 score: 0.767296
F2-Measure: 0.800525
ROC AUC: 0.962304
Legitimate Transactions Detected (True Negatives):  85247
Legitimate Transactions Incorrectly Detected (False Positives):  48
Fraudulent Transactions Missed (False Negatives):  26
Fraudulent Transactions Detected (True Positives):  122
Total Fraudulent Transactions in validation dataset:  148

Results for model with 5 weight for fraud class

calculate_and_plot_neural_network(model_5_weight,1,'Confusion Matrix with Weight 5','Class Weight 5','Neural Network')

Accuracy: 0.998982
Precision: 0.664865
Sensitivity AKA Recall: 0.831081
F1 score: 0.738739
F2-Measure: 0.791506
ROC AUC: 0.958601
Legitimate Transactions Detected (True Negatives):  85233
Legitimate Transactions Incorrectly Detected (False Positives):  62
Fraudulent Transactions Missed (False Negatives):  25
Fraudulent Transactions Detected (True Positives):  123
Total Fraudulent Transactions in validation dataset:  148

Results for model with 500 weight for fraud class

calculate_and_plot_neural_network(model_500_weight,2,'Confusion Matrix with Weight 500','Class Weight 500','Neural Network')

Accuracy: 0.985768
Precision: 0.097892
Sensitivity AKA Recall: 0.878378
F1 score: 0.176152
F2-Measure: 0.338542
ROC AUC: 0.972720
Legitimate Transactions Detected (True Negatives):  84097
Legitimate Transactions Incorrectly Detected (False Positives):  1198
Fraudulent Transactions Missed (False Negatives):  18
Fraudulent Transactions Detected (True Positives):  130
Total Fraudulent Transactions in validation dataset:  148

Summarizing the Results

import matplotlib.pyplot as plt

def plot_summary_barcharts():
    #colors=['blue', 'yellow', 'green']
    colors=['lightskyblue', 'cornflowerblue', 'blue']
    edgecolor="grey"
    label_font = {'family': 'serif', 'color':  'darkblue', 'weight': 'normal', 'size': 12 }
    title_font = {'family': 'serif', 'color':  'black', 'weight': 'normal', 'size': 18 }
    footer_font = {'family': 'serif', 'color':  'black', 'weight': 'normal', 'size': 15 }
    fig, (ax1, ax2, ax3) = plt.subplots(3, 1)
    fig.set_size_inches(15, 50) 
    
    #ax1 = fig.add_subplot(1,1,1)
    ax1.set_xlim(left=0.7, right=1)
    ax1.barh(np.arange(len(arr_roc_auc)), arr_roc_auc, align='center',edgecolor=edgecolor, color=colors )
    ax1.set_xlabel('AUC Value',fontdict=footer_font)
    ax1.set_yticks(np.arange(len(arr_roc_auc)))
    ax1.set_yticklabels(arr_barchart_labels,fontdict=label_font)
    ax1.set_title('AUC across Algoritms',fontdict=title_font)
    ax1.set_facecolor('w')

    #ax2 = fig.add_subplot(2,1,2)
    ax2.set_xlim(left=0.97, right=1)
    ax2.barh(np.arange(len(arr_accuracy)), arr_accuracy, align='center',edgecolor=edgecolor, color=colors )
    ax2.set_xlabel('Accuracy Value',fontdict=footer_font)
    ax2.set_yticks(np.arange(len(arr_accuracy)))
    ax2.set_yticklabels(arr_barchart_labels,fontdict=label_font)
    ax2.set_title('Accuracy across Algoritms',fontdict=title_font)
    ax2.set_facecolor('w')
    
    ax3.set_xlim(left=0.75, right=0.9)
    ax3.barh(np.arange(len(arr_recall)), arr_recall, align='center',edgecolor=edgecolor, color=colors )
    ax3.set_xlabel('Recall Value',fontdict=footer_font)
    ax3.set_yticks(np.arange(len(arr_recall)))
    ax3.set_yticklabels(arr_barchart_labels,fontdict=label_font)
    ax3.set_title('Sensitivity AKA Recall across Algoritms',fontdict=title_font)
    ax3.set_facecolor('w')
    
    plt.show()

plot_summary_barcharts()

def plot_summary_table():
    #results=pd.DataFrame(list(arr_table_labels,arr_accuracy,arr_roc_auc,arr_recall,arr_precision,arr_f1score,arr_f2measure,arr_tp,arr_fn,arr_tn,arr_fp))
    #results.columns = ['Algoritm','Accuracy','ROC-AUC','Recall','Precision','F1-Score','F2-Measure','TP','FN','TN','FP']
    
    results = pd.DataFrame({
    'Algoritm': arr_table_labels,
    'Accuracy': arr_accuracy,
    'ROC-AUC': arr_roc_auc,
    'Recall': arr_recall,
    'Precision': arr_precision,
    'F1-Score' : arr_f1score,
    'TP' : arr_tp,
    'FN' : arr_fn,
    'TN' : arr_tn,
    'FP' : arr_fp
    })
    
    return results

plot_summary_table()

	Algoritm	Accuracy	ROC-AUC	Recall	Precision	F1-Score	TP	FN	TN	FP
0	BAGGING -> Sampling Straegy : 0.01	0.999122	0.937494	0.831081	0.710983	0.766355	123	25	85245	50
1	BAGGING -> Sampling Straegy : 0.1	0.996781	0.945874	0.837838	0.330667	0.474187	124	24	85044	251
2	BAGGING -> Sampling Straegy : 1	0.975235	0.961538	0.871622	0.057951	0.108677	129	19	83198	2097
3	RANDOM FOREST -> Class Weight 1	0.999263	0.944639	0.824324	0.767296	0.794788	122	26	85258	37
4	RANDOM FOREST -> Class Weight 100	0.995681	0.957426	0.851351	0.266385	0.405797	126	22	84948	347
5	RANDOM FOREST -> Class Weight 1000	0.989490	0.966482	0.858108	0.126494	0.220486	127	21	84418	877
6	LOGISTIC REGRESSION -> Class Weight 1	0.994640	0.953544	0.844595	0.223214	0.353107	125	23	84860	435
7	LOGISTIC REGRESSION -> Class Weight 100	0.994640	0.953544	0.844595	0.223214	0.353107	125	23	84860	435
8	LOGISTIC REGRESSION -> Class Weight 1000	0.994640	0.953544	0.844595	0.223214	0.353107	125	23	84860	435
9	DECISION TREE -> Class Weight 1	0.999122	0.918620	0.837838	0.708571	0.767802	124	24	85244	51
10	DECISION TREE -> Class Weight 100	0.998912	0.905024	0.810811	0.648649	0.720721	120	28	85230	65
11	DECISION TREE -> Class Weight 1000	0.998876	0.905007	0.810811	0.638298	0.714286	120	28	85227	68
12	XGBOOST -> Class Weight 1	0.999567	0.970665	0.790541	0.951220	0.863469	117	31	85289	6
13	XGBOOST -> Class Weight 50	0.999146	0.974359	0.837838	0.716763	0.772586	124	24	85246	49
14	XGBOOST -> Class Weight 1000	0.998853	0.957629	0.837838	0.626263	0.716763	124	24	85221	74
15	NEURAL NETWORK -> Class Weight 1	0.999134	0.962304	0.824324	0.717647	0.767296	122	26	85247	48
16	NEURAL NETWORK -> Class Weight 5	0.998982	0.958601	0.831081	0.664865	0.738739	123	25	85233	62
17	NEURAL NETWORK -> Class Weight 500	0.985768	0.972720	0.878378	0.097892	0.176152	130	18	84097	1198